On 2017 Jul 23, Randi Pechacek commented:

Embriette Hyde made a blog post about this paper on microBEnet as part of a larger discussion on the microbiome of the built environment of our cars.

On 2014 Jun 19, Swapnil Hiremath commented:

This guideline was discussed on June 10th 2014 on the open online nephrology journal club, #NephJC, on twitter. Introductory comments are available on PBFluids and at the NephJC website. It was quite a spirited discussion, with participation from nephrologists, clinical pharmacologists, internists, and more. A transcript and three different curated (i.e. Storified) versions of the tweetchat are available at the same NephJC link.

On June17th 2014, we conducted a video chat via Google Hangout, among the NephJC editors, Dr Richard Sterns and Dr Hatim Hassan, an archived version of which can be viewed on Youtube.

The highlights of the tweetchat and the hangout were: 1. These guidelines are extensive and exhaustive and will serve as an extremely useful resource for students, residents and practicing physicians. 2. There was widespread agreement that 'asymptomatic' hyponatremia is rarely asymptomatic, and doing away with that qualifier is a good move. 3. The recommendation against use of vasopressin antagonists in chronic hyponatremia is appropriate given lack of superiority in comparison against standard treatment, and possibility of neurological sequelae from rapid correction (and the high cost of these agents remains a concern). 4. The empiric treatment with hypertonic saline in hyponatremic patients with moderate to severe symptoms will be quite handy, particularly since the intricate calculations otherwise needed are often found to be daunting. 5. The lack of strong evidence (made especially apparent by the use of the GRADE methodology) is disappointing, especially given how common hyponatremia is, and highlights a need for future research.

Interested individuals can track and join in the conversation by following @NephJC or #NephJC, or visit the webpage at NephJC.com.

This comment is cross-posted at the other two versions of the guidelines also.

On 2014 Feb 25, Patrick Morcillo commented:

Let me add a few points to clarify and avoid misunderstandings, mostly for the lay public: 1) There is no "placebo effect" in mice, and therefore, the results are real 2) When no voltage (or current) is applied, there is no effect 3) When the current is applied in a non-acupoint, there is no effect 4) Please be aware that many promising results in mice do not work with humans

Disclaimer: I am acknowledged in the article

On 2016 Aug 23, Ben Goldacre commented:

This trial has the wrong trial registry ID associated with it on PubMed: both in the XML on PubMed, and in the originating journal article. The ID given is NCT0024451. We believe the correct ID, which we have found by hand searching, is NCT00244517.

This comment is being posted as part of the OpenTrials.net project^[1] , an open database threading together all publicly accessible documents and data on each trial, globally. In the course of creating the database, and matching documents and data sources about trials from different locations, we have identified various anomalies in datasets such as PubMed, and in published papers. Alongside documenting the prevalence of problems, we are also attempting to correct these errors and anomalies wherever possible, by feeding back to the originators. We have corrected this data in the OpenTrials.net database; we hope that this trial’s text and metadata can also be corrected at source, in PubMed and in the accompanying paper.

Many thanks,

Jessica Fleminger, Ben Goldacre*

[1] Goldacre, B., Gray, J., 2016. OpenTrials: towards a collaborative open database of all available information on all clinical trials. Trials 17. doi:10.1186/s13063-016-1290-8 PMID: 27056367

* Dr Ben Goldacre BA MA MSc MBBS MRCPsych<br> Senior Clinical Research Fellow<br> ben.goldacre@phc.ox.ac.uk<br> www.ebmDataLab.net<br> Centre for Evidence Based Medicine<br> Department of Primary Care Health Sciences<br> University of Oxford<br> Radcliffe Observatory Quarter<br> Woodstock Road<br> Oxford OX2 6GG

On 2014 Jun 25, Ryan Radecki commented:

Post-publication commentary:

"Should Paramedics Intubate Out-of-Hospital Cardiac Arrest?"

Airway management of out-of-hospital cardiac arrest is a controversial topic. Most patients transported for OHCA have receive prehospital airway management. However, attempts at establishing an airway can interrupt compressions, over-ventilation can decrease cerebral perfusion, and delays in airway acquisition impact transport to definitive care....

http://www.emlitofnote.com/2014/06/should-paramedics-intubate-out-of.html

On 2014 Feb 25, Maciej Wójcik commented:

very nice work

On 2014 Feb 23, Hilda Bastian commented:

This paper tackles an important issue. We definitely need better ways to keep up with the evidence - and the rate of growth of that evidence makes it both more difficult and more urgent (Bastian H, 2010). It's particularly helpful that the paper addresses the risks of multiple testing in continuous updating models.

In calling for "a shift to continuous work process," though, it's important to remember that this shift has long occurred for many organizations and groups. A 2010 survey of agencies that sponsor and conduct systematic reviews (sometimes with clinical practice guidelines as well), found 66 that were already doing this to some extent at least (Garritty C, 2010).

In this latest proposal for living systematic reviews, several issues reach Table 1 as key challenges, that are unquestionably important. But "validation and acceptance by the academic community" and "ensuring conventional academic incentives are maintained" did not prevent the development of continuous updating models.

The restriction of access to key databases does contribute to keeping many groups trapped in duplicative updating hamster wheels, though. Poor access leads to critical research waste (Glasziou P, 2014). Making the preservation of conventional academic incentives foundational in Table 1, rather than, say, opening databases, runs the risk of focusing us on technical issues within restricted models, slowing down and limiting both innovation and the entry of new players.

On 2015 Dec 11, Mark Johnston commented:

An interactive protocol from this article is freely available at protocols.io.

On 2016 Aug 23, Ben Goldacre commented:

This trial has the wrong trial registry ID associated with it on PubMed: both in the XML on PubMed, and in the originating journal article. The ID given is NCT011583309. We believe the correct ID, which we have found by hand searching, is NCT01583309.

This comment is being posted as part of the OpenTrials.net project^[1] , an open database threading together all publicly accessible documents and data on each trial, globally. In the course of creating the database, and matching documents and data sources about trials from different locations, we have identified various anomalies in datasets such as PubMed, and in published papers. Alongside documenting the prevalence of problems, we are also attempting to correct these errors and anomalies wherever possible, by feeding back to the originators. We have corrected this data in the OpenTrials.net database; we hope that this trial’s text and metadata can also be corrected at source, in PubMed and in the accompanying paper.

Many thanks,

Jessica Fleminger, Ben Goldacre*

[1] Goldacre, B., Gray, J., 2016. OpenTrials: towards a collaborative open database of all available information on all clinical trials. Trials 17. doi:10.1186/s13063-016-1290-8 PMID: 27056367

* Dr Ben Goldacre BA MA MSc MBBS MRCPsych<br> Senior Clinical Research Fellow<br> ben.goldacre@phc.ox.ac.uk<br> www.ebmDataLab.net<br> Centre for Evidence Based Medicine<br> Department of Primary Care Health Sciences<br> University of Oxford<br> Radcliffe Observatory Quarter<br> Woodstock Road<br> Oxford OX2 6GG

On 2014 Mar 24, Karen Woolley commented:

FINALLY...more attention is being paid to PRACTICAL issues about data sharing!

Well done to Wilhelm, Oster, and Shoulson for reminding us that it takes resources (financial and nonfinancial) to share data. We should also remember that it takes resources (financial and nonfinancial) to publish data. This issue seems to have been lost in the hand-wringing taking place over low and slow publication rates.

A robust systematic review, presented at the 2013 Peer Review Congress (organised by JAMA and the BMJ) identified “lack of time” as the main reason why researchers don’t write up manuscripts.¹ Clearly, many researchers need writing support (ie, from legitimate, ethical, highly trained and qualified professional medical writers; NOT ghostwriters). Similar to Wilhelm et al., we outlined the need to consider the cost issue if we want to enhance publication speed and quality.² We think our paper struck a chord - it was among the top 5 most downloaded papers from Current Medical Research & Opinion that year.

Professional medical writers are trained to help make complex data understandable to different target audiences (eg, researchers, regulators, patients) and could, therefore, help address another critical point made by Wilhelm et al., “Standardization costs for data-sharing models include the additional effort required to share, beyond what is required of any high-quality clinical research, because it takes considerably more effort to organize and make data understandable to others.”

Wilhelm et al. conclude that “Understanding and planning for the costs [for data sharing] at the outset of research can help realize the full potential of data sharing.” The same sentence could apply to publications ie, understanding and planning for the costs of manuscript writing at the outset of research can help realise the full potential of peer-reviewed publications.

Authors and affiliations Karen L. Woolley PhD CMPP,a Art Gertel MS,b Cindy Hamilton PharmD,c Adam Jacobs PhD,d Jackie Marchington PhD CMPPe (Global Alliance of Publication Professionals; www.gappteam.org) a. Division Lead. ProScribe – Envision Pharma Group; Adjunct Professor, University of the Sunshine Coast, Australia. b. VP, Regulatory and Medical Affairs, TFS, Inc.. USA; Senior Research Fellow, CIRS. c. Assistant Clinical Professor, Virginia Commonwealth University School of Pharmacy; Principal, Hamilton House, USA. d. Director, Dianthus Medical Limited, UK. e. Director of Scientific Operations, Caudex Medical, UK.

Disclosures All authors declare that: (1) all authors have or do provide ethical medical writing services to academic, biotechnology, or pharmaceutical clients; (2) KW’s husband is also an employee of ProScribe – Envision Pharma Group; all other authors’ spouses, partners, or children have no financial relationships that may be relevant to the submitted work; and (3) all authors are active in national and international not-for-profit associations that encourage ethical medical writing practices. No external sponsors were involved in this study and no external funding was used.

References 1. Scherer RW, Ugarte-Gil C. Authors’ reasons for unpublished research presented at biomedical conferences: A systematic review. http://www.peerreviewcongress.org/abstracts_2013.html#1 Accessed 27 February 2014. 2. Woolley KL, Gertel A, Hamilton C, Jacobs A, Snyder G (GAPP). Poor compliance with reporting research results – we know it’s a problem…how do we fix it? Curr Med Res Opin 2012;28:1857-1860.

On 2014 Feb 21, Rafael Najmanovich commented:

For a in depth look at the method and its validation, take a look at Najmanovich R, 2008.

On 2014 Feb 27, guoan zhang commented:

axl and Tyro3 and Mer are important regulator of natural killer cell maturationCaraux A, 2006, and now those receptors are revealed to be players in the regulation of NK cells function in tumor immunology.it seems tumor ,through gas6 which could be secreted by tumor cells or stromal cells , suppressed the function of NK cells.and cbl-b takes a important role in gas6/TAM-mediated NK cells malfunction. and more important, the authors found a TAM kinase inhibitor and warfarin exerts anti-metastatic activity in mice via Cbl-b/TAM receptors in NK cells, suggesting possible clinical use.

On 2014 Feb 26, M S Muthu commented:

Another scientific vision from Prof Feng. Great....

On 2014 Nov 26, Guillaume Filion commented:

This article is one of the "CISCOM meta-analyses", which are very similar papers written by different authors. For more information about the CISCOM meta-analyses, check the blog post "A flurry of copycast on PubMed" at the following link http://blog.thegrandlocus.com/2014/10/a-flurry-of-copycats-on-pubmed

On 2014 May 23, Germana Bancone commented:

The authors state that the patient had a low G6PD enzymatic activity "The spectrophotometric assay showed a mildly reduced level of the enzyme: 7.2% (normal range: 8-18%)." Does 7.2% refer to the activity compared to a normal control? Could the authors specify the value expressed in international units per grams of hemoglobin (IU/gHb)? Could the authors explain this line "Though this X-linked disease is known to affect males exclusively, Errico et al., [2] have described it in females as well.", are they referring to G6PD deficiency or to hemolysis caused by ketoacidosis with G6PD deficient subjects?

On 2014 Nov 16, Ivan Oransky commented:

A medical resident explains how this paper came to be retracted: http://retractionwatch.com/2014/11/12/know-how-to-recognize-pseudoscience-whistleblower-fills-in-the-blanks-on-fish-oil-retraction/

On 2014 Apr 27, Jeff Kiefer commented:

The author seems to have mistakenly interpreted ref 80 Fantin VR, 2006. In the author's article he says, "Rats were fed fish oil or beef tallow." I could not find a mention of the diets that the animals used by Fantin et al were fed. In addition, the animals used in the Fantin VR, 2006 were not rats, they were FVB female mice. Lastly, in the authors discussion on Fantin VR, 2006, he states that mitochondrial enzyme activity was measured in kidney cells, which is wrong also. The mitochondrial experiments were performed on breast cancer cell lines. I doubt the veracity of the authors conclusions, as they relate to fish oil damaging mitochondria, do to the gross misrepresentation of Fantin VR, 2006 to support that particular hypothesis.

On 2016 Feb 18, Kristina Hanspers commented:

The pathway in figure 2 is available in the "Open Access Publication" collection at WikiPathways: http://www.wikipathways.org/index.php/Pathway:WP2637. This pathway can be downloaded for use in network analysis tools such as Cytoscape and PathVisio.

On 2014 Mar 01, L David Sibley commented:

It is interesting to see that most of the comments are about why we choose to study AMA1 and whether or not AMA1 is essential. We feel that there may be some misunderstanding about the rationale for our study, so hopefully the following points will help to clarify a few things.

The first misunderstanding deals with the findings of our previous work on ALD mutants and binding to the MIC tail (Starnes et al 2009). When we in a previous comment that these studies “were not able to definitively separate the functions of energy metabolism from motility vs. invasion”, we meant that while the studies did identify a mutant (TgALD K41E-R42G) that separates the functions of energy production from invasion, they did not fully explain why such mutants had no effect on gliding. One plausible explanation was provided in that paper: decreased occupancy of ALD-MIC2 in the cell (a product of altered affinity and protein concentrations) might differentially affect gliding vs. invasion. However, an alternative explanation is that ALD might also bind to another adhesin that is important in invasion but not gliding. When it became apparent that mutants in the AMA1 tails (AMA1t) also affect ALD binding in vitro (i.e. FW/AA) (Sheiner and Soldati et al, 2010), this provided a logical candidate, since AMA1 was known to be involved in host cell invasion but not motility (Mital et al., 2005).

Perhaps a second point of confusion is that our study did not try to address the essentiality of AMA1, something that has been studied by others, who ascribe various roles to the protein including attachment, MJ formation, and cell penetration. We were not aware at the outset of our study of the data contained in Bargieri et al. 2013, as the paper was published online while our work was under review. However, this likely would not have changed our approach as were not concerned with whether AMA1 is essential, or what factors might compensate for its loss, but rather with how it functions when it is expressed. Because the FW/AA mutations are located in the tail of the protein, we reasoned they were more likely to be involved in functions in the cytosol, rather than influencing the roles of the extracellular domains. Hence, this mutant provided an excellent candidate to test whether decreased cell invasion was due to alteration in ALD binding. As it turned out, study of additional mutants did not provide support for this model. Moreover, by re-examining the role of ALD in energy production, our study newly revealed that the previously ascribed role in adhesin binding does not play an essential role in vivo. We believe this is the only aspect of the apicomplexan invasion model that is directly addressed by our studies. We also hope that our work will inspire further studies to figure out the real function of AMA1t, which in turn will help us to understand the invasion process better.

Bang Shen, David Sibley

On 2014 Feb 27, Robert Menard commented:

The goal of the paper was to see whether aldolase plays a mechanical role during motility and/or internalization, the motor-dependent processes of the zoite. AMA1, using AMA1 KOs, was shown by the Bargieri paper (not cited in the current paper) to have NO detectable role in these processes, which would imply that the putative aldolase-AMA1 interaction does not either. The sound rationale appears to be that of the Starnes paper, which asked the question using the MIC2-aldolase interaction, since MIC2 is a bona fide motor-binding protein and is clearly important for motile processes. The data of the Starnes paper clearly indicate that aldolase has no mechanical role. For example, mutants K41A and K41E:R42G bind to MIC2t with 18% and 5% efficiency, respectively, and to actin with 13% and 9% efficiency, respectively. Fig 5C shows that these mutants have no detectable defect in motility. It is surprising to see that the author now says the Starnes data ‘was not able to definitely separate the functions’. This is clearly contrary to the straightforward title and abstract, which reads: “we generated a series of mutations in Toxoplasma gondii aldolase (TgALD1) that delineated MIC2 tail domain (MIC2t) binding function from its enzyme activity”.

Regarding the role of AMA1 in the TJ, the redundancy theory by AMA paralogs is presented incompletely. The author omits the crucial point that while AMA1 was not found to have any role in internalization at the TJ, it has an important role in adhesion to the host cell. AMA1 paralogs are expected to have a similar function, in promoting proper adhesion. Examination of Fig 1 shows overexpression of AMA1 paralogs in the AMA1 KO, but complete citation of the paper should also state increased paralog expression decreases the adhesion defect. In other words, AMA1 paralogs appear to be adhesins like AMA1, and are not expected to do at the TJ what AMA1 does not do itself. Lastly, we again point out that it is incorrect to state that inhibition experiments showed that the AMA1-RON interaction is important for TJ formation; they only indicate that its inhibition blocks invasion. This might occur by other scenarios than the interaction being the TJ, as further discussed in the Bargieri paper.

While we agree that it might be premature to draw a model of the TJ, it is equally dangerous to discard data that do not fit with the original model.

On 2014 Feb 26, L David Sibley commented:

As was discussed in our paper, and the previous Starnes paper, prior mutants of aldolase or MIC2t were not able to definitively separate the functions of energy metabolism from motility vs. invasion. Indeed, several mutants made in other systems have shown these later two functions can be independently controlled. Hence showing that one behavior is affected while the other one is not, cannot be used to argue that the process is not essential at some point. Combined with the hypothesis that AMAt binding to aldolase might be important for invasion (while playing no role in motility), this dichotomy provided more than adequate rationale for our present study. We also do not agree that prior work on AMA1 rules out a role in invasion, it merely shows that individual genes may be dispensable under some circumstances. Examination of Figure 1D in Bargierri et al., reveals evidence for 15-fold upregulation of a paralog of AMA1 (incorrectly called a homologue) in the TgAMA1 KO, suggesting that this gene may mask the phenotype by compensating for AMA1. Whether Plasmodium has recognizable paralogs of AMA1, or uses another adapter, one could still argue that the mechanism is conserved since the MJ is preserved. Independent of these genetic studies, there is a strong body of work that AMA1 is critical to the MJ formation. At this point it is premature to discard the existing model based on an incomplete assessment of potential compensatory mechanisms for loss of individual genes.

On 2014 Feb 25, Robert Menard commented:

The demonstration that aldolase plays no bridging role between actin and the MIC2/TRAP tails during zoite motility was provided by Starnes et al, who introduced mutations in aldolase that specifically abolished its capacity to bind MIC2 but not its energy producing capacity; these mutations had no effect on tachyzoite motility, which demonstrated the point. Regarding AMA1, previous work in Toxoplasma and Plasmodium has conclusively demonstrated not just that AMA1 is not essential for internalization, but that it has no detectable role in the process: AMA1 KO tachyzoites form a normal TJ and enter cells at a normal speed. The argument of redundancy by AMA1 paralogs has no basis, since (i) we showed that AMA1 - and paralogs - play a role in adhesion to the host cell, not directly in invasion at the TJ, and (ii) Plasmodium expresses no AMA1 paralog. These genetic data (Giovannini et al, CHM, 2011; Bargieri et al, Nat Comm, 2013) strongly argue against any motor-dependent function of AMA1 in any zoite and thus question the rationale of the present study. R. Menard, M. Meissner and D. Bargieri

On 2014 Feb 22, L David Sibley commented:

We welcome the opportunity to clarify the reasoning behind our study. We chose the AMA1-aldolase interaction to study because it demonstrates the requirement of two residues in the tail (FW) of AMA1 for cell invasion in a conditional knockdown, as reported previously (Sheiner et al., 2010 (PMID:2054586)). This AMA1t mutant also fails to bind aldolase in vitro, leading to the testable hypothesis that these two phenotypes are linked. We chose AMA1 not to address whether it is essential or not (indeed the jury is still out on this important question), but rather to test the model that binding of adhesin tails to aldolase is critical for linking the motor complex (as suggested previously by Jewett et al., 2003 (PMID12718875), and Starnes et al., 2009 (PMID19380114)). We are aware that others have questioned the requirement for AMA1 during invasion, but these studies also failed to account for possible redundant roles of paralogs, for which there is clear evidence of up-regulation in at least the one study that looked (Bargieri et al., Nature Comm 2013 (PMID24108241)). The use of knockout strains also runs the risk that suppressor mutations may have arisen in the background, given the length of time required to obtain such knockouts, at least by conventional means. In contrast, the conditional system we employed is less prone to such issues of compensation.

By studying a broader collection of AMAt mutants than previously, we show that there is no correlation between AMA1t-aldolase binding in vitro and invasion. This led us to further investigate the role of aldolase using more efficient techniques for gene disruption (Andenmatten et al., Nat Meth 2013 (PMID23263690)). The results show conclusively that aldolase is required for energy metabolism, but not binding to adhesin tails during invasion. This insight could not have been provided by any of the previous studies, as outlined in the discussion to our paper. Although we have not tested MIC2t, or other adhesins shown to bind aldolase, this no longer seems worthwhile given the clear lack of a phenotype for aldolase negative cells when grown in the absence of glucose. As we further point out in the discussion, there are clearly important roles for conserved residues in the tails of adhesins and the search is on for what those functions might be.

In terms of the model for invasion, our studies indicate that some other protein(s) must be responsible for linking the adhesin tails to the motor complex, and that if aldolase participates, it is redundant. This finding is consistent with the view that redundancy in biology is likely the norm for essential pathways. The work by Andenmatten et al., Nat Meth 2013 (PMID23263690) offers an exciting new tool to explore such redundancy with greater precision than previously possible, and indeed this was the basis for the final test in our study. Andenmatten et al., show that under conditions where genes can be rapidly excised by inducible Cre, it is possible to delete some of the core components of the glideosome. However, this is not without consequence as such mutants are severely impaired in gliding and invasion. Hence, we would interpret the available data as providing strong support for the current glideosome model for motility and invasion, rather than cause to abandon it. What remains unanswered by current studies is what is the backup or alternative mechanism(s), by which such deletion parasites still invade? There are a number of possibilities including: 1) residual levels of proteins remaining in knockout cells; 2) redundancy (i.e. paralogs that are upregulated), as are clearly evident in the genome, or 3) alternative mechanisms that serve as backup, albeit clearly less efficient. The remaining challenge for the field is to fill in these details, at which point it may be possible to provide a revised model for gliding and invasion by apicomplexans.

On 2014 Feb 20, Markus Meissner commented:

This nice study re-adresses the role of the glycolytic enzyme aldolase as a critical component of the gliding and invasion machinery of apicomplexan parasites as suggested previously by the same group. Aldolase has been described as a critical linker molecule between the tail domain of microcemal transmembrane proteins, such as AMA1 or MIC2 and the actin-myosin motor, thereby playing a crucial role for force transmission during motility and invasion. Here the authors tested if the interaction between AMA1 and ALD is important for host cell invasion. However, previous analysis of AMA1 knockdown (the same mutant as used in this study) and knockout mutants ruled out an important role of AMA1 as force transmitter during gliding motility (Mital et al., 2005) and host cell invasion (Giovannini et al., CHM 2011; Bargieri et al., Nature Comm 2013). Why did the authors expect a critical role of aldolase binding to AMA1, when AMA1 itself is not required for force transmission? Did the authors also investigate the role of MIC2 in binding to aldolase?

The authors continued to analyse the phenotype of an ALD knockout and demonstrate that these parasites can invade the host cell normally, ruling out an important function of aldolase during this process.

It would have been very informative for the reader if the authors would have discussed their finding in a more holistic view that also incorporates recent findings on the gliding and invasion machinery. The authors mention that the current model needs to be revised, which is certainly true, since several of the core components for invasion appear to be not essential for invasion, including actin itself (Andenmatten et al., Nat Meth 2013). It would have been very interesting to hear the opinion of the senior author how he currently perceives the molecular mechanisms involved in gliding and invasion. As it stands now it seems that we do not have a model that is sufficiently backed up by experimental data.

On 2014 Nov 05, Gary Ward commented:

This is a beautiful and clear demonstration of how Toxoplasma gondii can serve as both a useful model organism for the study of other apicomplexan parasites, and powerful surrogate system for small molecule screening. By complementing TgCDPK3 with Plasmodium falciparum CDPK1 (PfCDPK1), the group was able to confirm the functional localization dependence of PfCDPK1 and identify compounds that inhibit both PfCDPK1 and TgCDPK3, as well as those that inhibit PfCDPK1 alone. This work and the work of Sharling et al. PLOS Negl Trop Dis [2010] 4: e794 and others provide good examples of how studying T. gondii may be useful to understanding other apicomplexan parasites from a drug development standpoint.

Posted by Gary Ward on behalf of the University of Vermont Toxoplasma Journal Club (UVM ToxoJC); members include Sam Ashley, Jenna Foderaro, Anne Kelsen, Shruthi Krishnamurthy, Jacqueline Leung, Pramod Rompikuntal & Gary Ward

On 2014 Feb 26, David Keller commented:

Why is citalopram still being used instead of safer escitalopram?

In the CitAD trial, the use of citalopram predictably led to its known adverse side-effect of QT-interval prolongation, which is a dose-dependent risk factor for toursades de pointes, an often-fatal cardiac arrhythmia. The QT-interval is prolonged further by interactions with numerous common medications, including over-the-counter omeprazole, which the patient may take or be prescribed inadvertently (1). As I have pointed out before (2), (3), the use of escitalopram instead of citalopram reduces the amount of QT-interval prolongation for any given degree of intended therapeutic antidepressant or antianxiety effect. The reason is that the therapeutic benefits of racemic citalopram are caused only by the levorotatory optical isomer (S-citalopram or escitalopram). The dextrorotatory molecule (R-citalopram) causes a roughly equivalent degree of QT-interval prolongation as escitalopram without contributing any known therapeutic effects. The last remaining reason to prescribe citalopram vanished when escitalopram went generic.

Are the cognitive adverse effects of citalopram caused roughly equally by both optical isomers, similar to the cardiac risks? If so, then the significant benefits of citalopram on agitation in Alzheimer's disease could have been achieved with roughly half the degree of cognitive impairment by substituting escitalopram at half the milligram dose instead of citalopram.

(1) Yee Guan Yap, A John Camm. Drug induced QT prolongation and torsades de pointes. Heart. 2003 November; 89(11): 1363–1372.PMCID: PMC1767957

(2) Keller DL. Prescribe escitalopram instead of citalopram. Am J Med. 2013 Jun;126(6):e21. doi: 10.1016/j.amjmed.2012.10.024. No abstract available. PMID: 23684405 [PubMed - indexed for MEDLINE]

(3) Keller DL. Comments and questions regarding the safety of citalopram. Mayo Clin Proc. 2013 Apr;88(4):420. doi: 10.1016/j.mayocp.2013.01.023. No abstract available. PMID: 23541017 [PubMed - indexed for MEDLINE]

On 2014 Nov 11, Eva Kottenberg commented:

Contrary to his statement, we have not been contacted by Dr. Berthelsen from Denmark. Also, contrary to his statement, ethics approval was explicitly mentioned in our paper (page 454 line 52). We had also clearly and explicitly reported in our paper, that our study is a retrospective analysis of a subgroup in an ongoing trial. Finally, we had specifically reported in which respect the patient selection of the current analysis differed from that of our prior Lancet paper. Dr. Berthelsen could have avoided his mathematical speculations simply by reading our paper much more carefully, so as to avoid obvious false statements which question the integrity and honesty of our scientific work. PD Dr. med. E. Kottenberg, Prof. Dr. med. J. Peters

On 2014 Oct 29, Eva Kottenberg commented:

None

On 2014 Oct 22, Preben Berthelsen commented:

Before accepting the authors’ results it must be realized that the conclusions are based on a non-randomized, unplanned, post hoc subgroup analysis of a larger study on remote ischaemic preconditioning in CABG surgery (ClinicalTrials NCT01406678). The results of the primary study were published in The Lancet (August 17, 2013). There are severe methodological problems with The Lancet paper as can be seen in the PubMed Commons comment to the paper (PMID:23953384).

In the present paper, the authors have selected, as a control group, 130 patients of the 167 controls included in the original Lancet paper. The patients were anaesthetized with isoflurane and no remote ischaemic preconditioning was used. The results are peculiar. The average 72h troponin release AUC in the original 167 patients was 321 (SD 213) and in the present subgroup of 130 patients 514 (SD 600). It is a mathematical impossibility that that so large a difference - in both mean and SD values - can be correct when 78% of the patients/results are shared. Furthermore, in the present study 14 of 130 (11%) are reported to be ACE/ARB treated while 75 of 167 (45%) in The Lancet paper are so treated. Again, it is not mathematically possible that the reported numbers are correct. I have tried to contact the corresponding author twice to determine if these discrepancies are printing errors. I have not received a response.

The authors have not statistically compared the difference in troponin release between sulphonylurea-treated diabetics and patients without diabetes. Their conclusions are instead solely based on within-group statistical analyses. And as Bland & Altman lucidly put it “this approach is biased and invalid, producing conclusions which are potentially highly misleading” (Trials 2011,12:264).

The investigation has no approval from an ethics committee. Taken in all, I feel it justified to view the results of this paper with scepticism. P.G.Berthelsen, Charlottenlund, Denmark.

On 2014 Jul 29, Jesper M Kivelä commented:

1) Under the Author information and after Children's Hospital there should be Helsinki University Central Hospital and University of Helsinki, Helsinki, Finland.

2) My academic degree, in addition to MD, is not PhD as indicated in Wiley Online Library under Author Information. I'm a PhD student at the moment.

On 2016 Jan 09, Chao Liu commented:

Findings from our studies are consistent with the authors’ point. We found that glucocorticoids could promote renal water and sodium excretion in heart failure.

1. Liu C, Chen Y, Kang Y, Ni Z, Xiu H, Guan J, et al. (2011). Glucocorticoids Improve Renal Responsiveness to Atrial Natriuretic Peptide by Up-Regulating Natriuretic Peptide Receptor-A Expression in the Renal Inner Medullary Collecting Duct in Decompensated Heart Failure. J Pharmacol Exp Ther 339(1): 203-209.

2. Liu C, Liu G, Zhou C, Ji Z, Zhen Y, Liu K (2007). Potent diuretic effects of prednisone in heart failure patients with refractory diuretic resistance. Can J Cardiol 23(11): 865-868.

3. Liu C, Liu K (2014a). Effects of glucocorticoids in potentiating diuresis in heart failure patients with diuretic resistance. Journal of cardiac failure 20(9): 625-629.

4. Liu C, Liu K (2014b). Reply to Day et al.--hypouricemic effect of prednisone in heart failure: possible mechanisms. Can J Cardiol 30(3): 376 e373.

5. Liu C, Zhao Q, Zhen Y, Gao Y, Tian L, Wang L, et al. (2013). Prednisone in Uric Acid lowering in Symptomatic Heart Failure Patients With Hyperuricemia (PUSH-PATH) study. Can J Cardiol 29(9): 1048-1054.

6. Liu C, Zhao Q, Zhen Y, Zhai J, Liu G, Zheng M, et al. (2015). Effect of Corticosteroid on Renal Water and Sodium Excretion in Symptomatic Heart Failure: Prednisone for Renal Function Improvement Evaluation Study. J Cardiovasc Pharmacol 66(3): 316-322.

7. Meng H, Liu G, Zhai J, Zhen Y, Zhao Q, Zheng M, et al. (2015). Prednisone in Uric Acid Lowering in Symptomatic Heart Failure Patients with Hyperuricemia - The PUSH-PATH3 Study. The Journal of rheumatology 42(5): 866-869.

On 2014 Nov 26, Guillaume Filion commented:

This article is one of the "CISCOM meta-analyses", which are very similar papers written by different authors. For more information about the CISCOM meta-analyses, check the blog post "A flurry of copycast on PubMed" at the following link http://blog.thegrandlocus.com/2014/10/a-flurry-of-copycats-on-pubmed

On 2014 Mar 01, Christopher Southan commented:

Source updates related to this paper and the availabilty of UniProt cross-references as PMC outlinks or suplementary data are detailed at http://cdsouthan.blogspot.se/2014/02/getting-to-know-databases-by-comparing.html. Our earlier paper mentioned in the abstract above, is http://www.ncbi.nlm.nih.gov/pubmed/22821596

Nov 2015 contextual comments added to Kudos http://goo.gl/OlK8xM

On 2015 Jan 24, Madhusudana Girija Sanal commented:

How cancer stem cells may originate-the three hit hypothesis-the importance of epigenetic hit!

There are thousands (if not millions) of dangerous cancer associated mutations in every "healthy" human being. However, only a few get cancer. First of all one mutation may not lead to cancer, but two or more hits (the classic two hit hypothesis) might. What happens is: (A) a genetic mutation (usually associated with proliferation) (B) another genetic mutation (associated with proliferation, cell attachment, a protein regulating the epigenetic status etc.) (C) Epigenetic instability or change (induced by another mutation or by the environment-mechanical, transcription factor induced or cytokine/growth factor induced). However, some mutations are very strongly proliferative/oncogenic that an epigenetic change is seldom required during the cancer initiation.

The events can be in any direction- ABC, CBA, CAB. Each of these has clinical examples, although there could be significant overlap. Esophagial cancer (some forms) is an example of CAB. Persistent acid reflux from the stomach (GERD) will induce the esophagial epithelium to undergo metaplasia. A transdifferentiation to columnar cells. May be columnar cells are more resistant to acid reflux or it is a futile attempt of our body to resist acidity. However, the transdifferentiation requires an epigenetic change from the stratified squamous epithelium epigenome to columnar cell epigenome which involves activation or repression of several transcription factors. This weakens the stability of epigenomic landscape (which is unique to each cell type). At this point, if an oncogenic mutation occurs, in any of these dysplastic cells it may lead to cancer. But it is possible that an oncogenic mutation pre-exists in these cells and the "loose" epigenetic atmosphere now helped the oncogene to express taking over the normal cell cycle which is followed by another mutation which helped the cells to invade (ACB). Now at this point which cell one would call a cancer stem cell? Probably, it is possible that the very cell which initiated the cancer may not express even a single cancer stem cell marker! These stem cell markers may come and go at a later stage and they are dictated by the tumor environment. In short, cancer stem cell markers are the need of the time because, at some point of cancer evolution they help cancer cells to adapt and survive. In short cancer cells, which are not cancer stem cells can give rise to a cancer stem cells and cancer stem cells can give rise to cancer cells which are not qualified to be called cancer stem cells. Therefore, to conclude, cells which have a "loose" epigenetic status (on a background of one or more 'oncogene' mutations) are the ideal candidates for cancer "stem" cells. It is an epigenetically dynamic state. It is epigenetic 'anarchy' and 'promiscuity' for survival as a cancer cell.

On 2015 Nov 25, Thomas Cleveland commented:

Very nice work, and the associated software they developed, "WillItFit," has well-organized code that is relatively easy for an outsider to understand and modify. This is the best work I've seen so far, in terms of analyzing nanodiscs by small-angle scattering (not to disparage others---there is a large literature and I've only recently started exploring it).

On 2014 Oct 18, Pavel Baranov commented:

What is the difference between Open Reading Frame (ORF) and Coding Sequence (CDS)?

Thank you for the reply. I think the disagreement lies in our understanding of what Open Reading Frame is.

A simple and effective definition of ORF is a sequence of codons not interrupted with stop codons: nucleotide sequence is open for reading in one of the three (for RNA) or six (dsRNA) frames. ORF is an abstract notion, it can be found in any sequence. Both protein-codng and non-coding sequences have ORFs.

Coding Sequence (CDS) is the part of RNA that encodes protein. CDS often, but not always (exceptions are ribosomal frameshifting, stop codon readthrough, etc.), is located within a single ORF. A single ORF may contain more than one Coding Region if, for example, translation begins at different start codons within the same ORF.

From your reply I presume that you suggest to define ORF as a sequence of codons from start to stop (like CDS). But the problem with this detention is that it is unclear what should we consider as a start of ORF. AUG? But not all AUGs are starts and not all starts are AUGs. Also there are no starts in non-coding RNAs, but there are ORFs in non-coding sequences.

Besides if we use this definition, all eukaryotic mRNAs coding for multiple protein isoforms would need to be described as bi- or even polycistronic.

I hope that this discussion brings some clarity to the terminology used or at least it draws attention to a potential confusion when terms such as ORF, CDS and cistron are not explicitly defined.

On 2014 Oct 03, Jonathan C Kagan commented:

In response to the comment by Baranov, we would first like to thank you for your interest in our work. We agree that alternatively translated eukaryotic proteins most commonly share a reading frame, as our study mentioned. In fact, we discuss this point in the analysis of our ribosomal profiling data. For instance, we observed that truncations are more common than internal out-of-frame translation products. Regardless of whether the products share a reading frame, however, a point of interest is the regulation allowing ribosomes to initiate translation at more than one location on a transcript. We chose to use the terms polycistronic and bicistronic for two reasons. First, bicistronic mRNAs are operationally defined as transcripts that produce two stable proteins of distinct functions, regardless of whether the two proteins share sequence similarity. Our study clearly demonstrated that this is the case with the MAVS transcript. Whether these functionally distinct proteins share coding sequence with each other is functionally irrelevant. Second, a transcript producing a truncated protein (such as MAVS) conforms to generally accepted definitions of a polycistronic transcript, such as you provided: “…protein products of distinct coding ORFs are translated from the same transcript.” While the two MAVS proteins are encoded in the same reading frame, their production is initiated at unique start sites; thus, they have distinct coding sequences. In short, we consider them to be distinct ORFs that share a reading frame and therefore produce two proteins from a bicistronic mRNA.

On 2014 Aug 06, Pavel Baranov commented:

What is a polycistronic mRNA?

This is an interesting research article that provides insights into distinct functions of two proteoforms that differ at their N-termini due to the use of alternative translation initiation starts. Terming the corresponding mRNA bicistronic is, however, somewhat misleading. When we refer to a bacterial mRNA as polycistronic we imply that protein products of distinct coding ORFs are translated from the same transcript. Here the ORF is the same, but the translation initiates at different codons of that ORF. There are several examples of human proteins with truncated or extended N-termini produced from alternative translation initiation starts, see a recently discovered extension of PTEN for a startling example, see Hopkins BD, 2013. True bicistronic mRNAs are rare. Molybdopterin subunits MOCS2A and MOCS2B were reported to be produced from two different but overlapping ORFs at the same mRNA, see Stallmeyer B, 1999. If this were true the corresponding mRNA could be classified as bicistronic, however, further evidence suggested that these proteins are produced from alternatively spliced transcript variants, see Hahnewald R, 2006. A good example of an eukaryotic polycistronic mRNA could be found in Drosophilla tal (aka pri) mRNA that codes for four peptides produced from four distinct ORFs, see Kondo T, 2010. Perhaps, human mRNAs with uORFs encoding functional proteins should also be classified as polycistronic, but not mRNAs that encode multiple protein isoforms translated from the same ORF.

On 2014 Feb 20, Anne Marie Cunningham commented:

Pre-publication version of this commentary can be found here

On 2017 May 10, Stefan Holubar MD, MS commented:

A classic paper by Dr. Salvati that many of my more seasoned colleagues often refer to when discussing palliative treatment of low rectal cancer using serial electrocoagulation.

On 2017 May 10, Stefan Holubar MD, MS commented:

None

On 2016 Oct 06, Anna Gorczyca commented:

This is great work that you all have done. Did you look at effect modification by BMI classifications?

On 2014 Feb 28, Jessie Tenenbaum commented:

This is a great paper for people like me who think about p-values, and yet are by no means statisticians- highly recommended.

I've been trying wrap my head about the fact that p-value does NOT mean the likelihood my hypothesis is correct. Here's what I've come up with:

I could hypothesize that a given male subject has only Y-containing sperm. We could then do the experiment of having him mate 5 times. If all 5 progeny come out as male, the p-value is under .05. That is, there is less than 5% chance those results could be observed by random chance. BUT that does NOT mean there is less than 5% chance that I am wrong, because it was a "long shot" (to use the article's phrase) to begin with.

Does that seem right? Any other examples that would better illustrate this point?

On 2016 Jun 02, David C. Norris commented:

Kudos to Ms. S., the patient in Dr Rosenbaum’s opening vignette, for having the good sense to dismiss the clinical equivalent of a cheesy pick-up line, "What do you think is the number-one killer of women?" Far from demonstrating that Ms. S's "sense of risk was clearly less about fact than about feeling," her rejection of this population statistic as a decision input may well reflect precisely the opposite. Indeed, the more rational Ms. S's approach to her health care decision-making, the more irrelevant Dr Rosenbaum's population statistics would have seemed to her. In a consultation with a subspecialist, Ms. S. might rightly have hoped the 'denominator' would be 1. She might have hoped for a personalized assessment, conveyed to her in meaningful forms capable of supporting her rational deliberation about her choices, and her rational commitment to those choices.

'Fact resistance' should be regarded as a diagnosis of exclusion. Leaping to this diagnosis undermines effort to fortify the scientific grounds of communication with patients, much as leaping to diagnose malingering or somatizing devitalizes diagnostic effort.

On 2014 Mar 14, David Keller commented:

Regarding the New England Journal of Medicine's editorial concerning 23andMe and the FDA

The two most important questions regarding a direct-to-consumer commercial genetic testing lab (such as 23andMe) are:

1) Analytical validity: does this lab report accurate raw genetic test data ?

2) Clinical validity: does this lab provide accurate assessment of the statistical probability of disease associated with the raw genetic data ?

Analytical and clinical validity are required of any genetic test, and both should be enforced by the FDA. Has 23andMe been accused of any specific cases of reporting erroneous raw genetic test data to consumers? Has 23andMe been accused of reporting erroneous statistical probabilities of disease associated with the genetic data of any patient? No such allegations appear in this editorial.

The FDA also expressed concerned about confused individuals misusing their genetic test results, such as by inappropriately adjusting their own warfarin dose or requesting mastectomy. These examples of harmful outcomes are improbable and unrealistic because of the safeguards in the medical care system. Patients must obtain warfarin prescriptions from clinicians, whose duty it is to explain the need to closely monitor the INR anti-coagulation test regardless of a patient’s genetic profile. Similarly, it is absurd to believe that any surgeon would perform a mastectomy based on a single saliva sample. There would be layers of confirmatory testing first.

The editorialists predict approvingly that, within a decade, “a majority of health plans will make it easy for their members to have their genomes sequenced...with or without the help of their physicians” but that this goal will first “require a massive data bank of genome reference materials”. They fail to explain how progress toward these goals can be maintained despite the FDA’s actions against 23andMe, for which the editorialists also express approval.

I propose that the FDA conduct confirmatory testing of genetic samples, in order to test the analytical validity of 23andMe and other genetic test labs. This should be done at the expense of the commercial labs. The FDA should also confirm the clinical validity of the genetic tests by statistical analysis of the association between variations in raw genetic data and the probability of disease, as included in the lab’s reports to patients and physicians.

Finally, no commercial testing lab or medical equipment manufacturer should be held liable for the unauthorized misuse of their product or service, provided that adequate warnings have been supplied to the consumers involved.

On 2014 Feb 14, Bruno Ramalho Carvalho commented:

Since the association between assisted reproductive techniques (ART) and birth defects was firstly raised, controversies have been frequently presented in literature. This is an interesting study, but, in my opinion, no definite conclusions can be taken with current knowledge.

It is a fact: large studies with robust methodologies have suggested that children born after ART have an increased risk of birth defects compared with naturally conceived ones. According to literature, risk of major malformations (conditions that cause functional impairment or require surgical correction) may be increased in up to 40% after IVF/ICSI.

In a recent meta-analysis, Wen et al (2012) compared 124,468 children conceived by ART with spontaneously conceived children, and suggested a significantly increased risk of birth defects in the first group. However, the relative risk (RR=1.37, 95%CI 1.26-1.48) was lower than the suggestion of this study, with no increased risk for ICSI when compared with conventional IVF.

As a matter of fact, the recent evaluation of more than 15,000 children born after ART demonstrated similar birth defect rates comparing with naturally conceived children (Yan et al, 2011). A large population study in Denmark compared congenital abnormality rates among naturally conceived and fertility treatment offspring from subfertile couples, and found no differences in overall prevalence of congenital malformations. Also, this study indicated that parental factors, like increasing time to pregnancy, should be considerably associated with a greater risk of birth defects (Zhu et al, 2006), which is at least plausible and has been proposed by other authors (Seggers et al, 2012).

Finally, mounting evidence suggests that parental infertility may be an important independent risk factor for birth defects. It is noticeable that, in majority of studies, naturally conceiving mothers are significantly younger, more likely to be parous and more ethnically diverse than ART mothers, and attempts to stratify patients according to infertility history or paternal age, for example, are infrequent. Then, as suggested by Basatemur & Sutcliffe (2008), this is why it would not be wrong to consider the risk of birth defects more associated with heredity that with ART themselves.

It would be great to read authors' and other researchers' comments on what I've mentioned above.

On 2014 Feb 18, Jean-Jacques Letesson commented:

Please correct" Brucella strains are intracellular pathogens that belong to the β-2 proteobacteria group" by Brucella strains are intracellular pathogens that belong to the alpha-2 proteobacteria group.

Can somebody please give some information about the so called B. abortus A19(which biovar, method of attenuation ..etc) and at least some reference (in english) describing the original strain, its attenuation and its characterization.

On 2014 Apr 09, SATISH KALHAN commented:

The interesting and provocative paper by Wang et al is significant contribution to our understanding of the physiological adaptation to preterm birth and the potential mechanism/s of the development of insulin resistance during adult life in the prematurely born infants. By using complex statistical analysis, and adjusting for pertinent perinatal variables, the authors show a strong negative correlation between cord blood insulin levels and gestational age, and tracking of plasma insulin levels from birth to early childhood. Although the authors do not discuss the biological mechanism/s for their observations, these data raise some key questions: 1. Are these data only applicable to the black and Hispanic populations? Over 75% of the study population was minority with higher incidence of obesity and insulin resistance. Examination of the data separately for the black and Hispanic group may have been useful. 2. Plasma levels of insulin respond rapidly to nutrients, are modified by the metabolic milieu and by changes in other hormones. Although the authors discuss the possible impact of such changes on the insulin levels in childhood, they ignored the impact of maternal milieu and her clinical care during labor and delivery on the cord blood insulin levels. In addition, mothers of preterm babies had higher incidence of smoking, diabetes and pregnancy related illness. The inclusion of obese subjects (BMI over 30) added additional variable to these measurements. It is interesting that all babies born before 32 weeks were classified as appropriate for gestational age. 3. The insulin tracking data are the most interesting and show that the babies with high insulin at birth also had high insulin during childhood. Since the insulin levels were not measured in the “basal” state these data show that babies who responded with higher insulin level at birth continue to be high insulin responders in childhood The study by Wang et al is laudatory for its execution and statistical analysis. However relating the levels of a substrate or hormone, that is acutely responsive to nutritional and metabolic influences, in this instance insulin, to multisystem disorder such as obesity or type 2 diabetes or body weight which are cumulative effects of a number of variables over time, although statistically feasible, may not give us useful biological insights. The present data do not exclude the possibility that interruption of pregnancy prematurely caused a metabolic insult to developing pancreas that programs the babies to develop long term consequences.

On 2014 May 27, Keon Wook Kang commented:

0Thanks for your careful suggestion. We will use primary cultured HUVEC for further studies.

On 2014 May 21, Amanda Capes-Davis commented:

It is important to know that a cell line may not come from the expected tissue or cell type. In this case, ECV-304 is known to be cross-contaminated with T-24, a bladder carcinoma cell line. The authenticity of any human cell line can be checked by STR profiling. For a database of known cross-contaminated cell lines, see http://iclac.org/databases/cross-contaminations/.

On 2014 Mar 09, Mao Mao commented:

We also identified another HIP1-ALK fusion in a Chinese cohort (99 cases), http://www.ncbi.nlm.nih.gov/pubmed/24496003.

On 2014 May 14, Jim Woodgett commented:

Two points to consider: 1. Canonical Wnt signaling is not appreciatively activated upon genetic deletion of GSK-3beta due to the redundant action of GSK-3alpha (Axin associates with either isoform). See: PMID: 17543867 2. 6-bromoindirubin 3'-oxime (BIO) is not isoform selective, nor are any other small molecule GSK-3 inhibitors. Hence, it is highly likely that the observed results on expansion of hematopoietic progenitors (also previously observed in animals treated with lithium: PMID: 16341242) are due to inhibition of GSK-3 (rather than GSK-3beta alone).

On 2017 Jul 01, David Keller commented:

Please contact Dr. Pedley or Dr. Fisch for copies of this paper [1].

I am not authorized to distribute copies.

1: Fisch BJ, Pedley TA, Keller DL. A topographic background symmetry display for comparison with routine EEG. Electroencephalography and clinical neurophysiology. 1988; 69(5):491-4. PubMed [journal] PMID: 2451597

On 2014 Mar 08, David Reardon commented:

Findings in Doubt Due to Failure to Account for Interrelationships Between Breast Cancer, Smoking & Abortion

Like similar studies exploring the association between smoking and breast cancer, this study by Kawai et al¹ unfortunately fails to explore a very important and intertwined risk factor: abortion history.

Numerous studies have shown that young women report starting or smoking more in order to cope with feelings associated with past abortions.

For example, a very recent longitudinal study published in the Journal of Adolescent Health revealed young women with a history of abortion had adjusted higher 4.1 times higher risk of smoking (CI, 1.9-8.8) and 4.5 times higher risk of nicotine dependence (OR 4.5; CI, 2.1-9.6).² Similar results are reported by others.³(4) In one post-abortion follow-up study, nearly one fourth of the women specifically described that they used smoking to "deal with" feelings related to their abortions.⁵

The importance of examining the three-way associations between smoking, abortion history and breast cancer is underscored by the controversy regarding statistical associations between abortion history and breast cancer. That controversy was recently reignited by meta-analysis of 36 studies conducted in China which found a significant association between abortion and breast cancer, including a dose effect. ⁶

To my knowledge, while plenty of researchers have explored the associations between smoking and breast cancer and abortion and breast cancer, none have yet to look at both risk factors in the same study. This is a serious problem, especially if one of these factors is actually just a proxy for the other.

Clearly, whether smoking and abortion arise from common risk factors or from causal interactions, the fact that they are associated in any fashion raises important research questions:

• Is the elevated risk of breast cancer associated with abortion due to behavioral changes (such as increased smoking) with the biological mechanism behind the elevated breast cancer rate due to smoking? or

• Is the apparent elevated risk of breast cancer associated with smoking really due to increased exposure to abortion in the population of smoking women and it is abortion (perhaps due to disruption of early pregnancy hormone cycles) contributing a biological mechanism that accounts for all or part of the observed increased cancer risk associated with smoking?, or

• Is there a combination of incidental associations and/or overlapping biological risk factors?

In my view, it clear that new analyses must be done which, when looking at the abortion history variable, segregate smokers from non-smokers. This would show if abortion has an independent effect. Similarly, when looking at the smoking history, the analyses should include segregation of smokers and non-smokers relative to history of 0, 1, or 2+ abortions.

It is my hope that Dr. Kawai's team, and others with similar data sets, will begin to explore these interactions.

References

(1) Kawai M, Malone KE, Tang MT, Li CI. Active smoking and the risk of estrogen receptor-positive and triple-negative breast cancer among women ages 20 to 44 years. Cancer. 2014 Feb 10. doi: 10.1002/cncr.28402.

(2) Olsson CA, Horwill E, Moore E, Eisenberg ME, Venn A, O'Loughlin C, Patton GC. Social and Emotional Adjustment Following Early Pregnancy in Young Australian Women: A Comparison of Those Who Terminate, Miscarry, or Complete Pregnancy. J Adolesc Health. 2014 Jan 15. pii: S1054-139X(13)00738-6. doi: 10.1016/j.jadohealth.2013.10.203.

(3) Pedersen, W. Childbirth, abortion and subsequent substance use in young women: a population-based longitudinal study. Addiction. 2007 102: 1971–1978.

(4) L Henriet, M Kaminski. Impact of induced abortions on subsequent pregnancy outcome: the 1995 French national perinatal pregnancy survey, Br J Obstet Gynaecol 2001 108:1036-1042.

(5) Major B, Richards C, Cooper ML et al. Personal resilience, cognitive appraisals, and coping: An integrative model of adjustment to abortion. J Person Soc Psychol, 1998; 74: 735-752.

(6) Huang Y1, Zhang X, Li W, Song F, Dai H, Wang J, Gao Y, Liu X, Chen C, Yan Y, Wang Y, Chen K. A meta-analysis of the association between induced abortion and breast cancer risk among Chinese females. Cancer Causes Control. 2014 Feb;25(2):227-36. doi: 10.1007/s10552-013-0325-7. Epub 2013 Nov 24.

On 2014 Feb 19, Francesca Demichelis commented:

Thank you for your comment. The yearly brachytherapy accrual rate at the study site is ~50. The number of patients included in the study was limited by tissue availability for biomarker assays. The 2000-2008 average postplanning values for V100 and D90 (at one month) were 93% and 158 Gy, respectively. Therefore, it is the molecular stratification of prostate cancer in these patients which may explain the high failure rate in a predominantly low-risk population, not the dosimetry. The significance of the molecular alterations in this setting needs to be tested by independent studies.

On 2014 Feb 12, Wayne Butler commented:

The regression did not account for prostate dosimetric quality parameters such as D90 and V100. An 11% failure rate in a predominantly low-risk population is very high, so poor quality implant dosimetry is probably a major factor. Their accrual rate of about 10 brachytherapy patients per year is well below the threshold necessary to become proficient in the operative procedure. It would be simple to compare mean dosimetry parameters between failures and non-failures, but that crucial data was not part of the analysis, so it remains unknown whether molecular alterations would remain as an independent predictor.

On 2014 Feb 20, David Keller commented:

If the patient requested a consultation with a genetic counselor, I would refer him to one. If he requested me to explain his results to him, I would do so based on my study of his genetic profile. As a 23andMe customer, I found their reports to be well-written and easy to interpret, and I would have no problem explaining the results of such a report. I rely on the FDA to ensure the accuracy of the raw test data, and the accuracy of its association with increased or decreased probability of disease. Beyond that, I favor maximizing the degree of control and autonomy patients can exert over their own health care, within reason. This philosophy demands that competent adults behave responsibly. Companies which supply tests to consumers should not be held responsible when individuals misuse their test results or violate an agreement to discuss their results with their physician before acting on them. If we demand that the government protect us from our own failure to act responsibly, then all direct-to-consumer testing companies will be regulated out of existence, and we will be left with fewer choices.

On 2014 Feb 19, John French commented:

In regard to the FDA's closure of a personal genomic service, I take the main issues of concern to the writer, an internist, is the accuracy of 23andMe's reported conclusion based upon the statistical association of the variant in question for PD. There is very little information on both type 1 and type 2 error rates on test accuracy which calls into question the reported statistical association. PD and most other diseases are polygenic with tens if not hundreds of variants genome contributing to varying levels of cumulative risk along with other contributing intrinsic and extrinsic factors. A main concern of the FDA was in regard to the lack of published validation studies for the methodologies used by this and other personal genomic services. Your response to a hypothetical patient seems appropriate. If they did not include such advice in their report to you, 23andMe should have offered similar caveats. Without sufficient characterization of significant intrinsic and extrinsic factors for the individuals in the candidate gene or genome wide association studies, the data cannot be weighted and is inconclusive. Where are the genetic counselors? Would you refer your patient to a genetic counselor?

Yandell et al. A probabilistic disease-gene finder for personal genomes. Genome Res.21(9): 1529–1542, 2011. doi: 10.1101/gr.123158.111 Elizabeth T. Cirulli & David B. Goldstein. Uncovering the roles of rare variants in common disease through whole-genome sequencing. Nature Reviews Genetics 11, 415–30, 2010) doi:10.1038/nrg2779

On 2014 Feb 17, David Keller commented:

Why I Care About the FDA's Closure of 23andMe's Personal Genomic Service

I am an internist and a 23andMe customer. I paid $25 for their genetic test kit several years ago as part of their campaign to identify a cohort of Parkinson disease (PD) patients carrying the LRRK-2 gene. In return, I received an extensive genetic report, as part of which I learned that LRRK-2 is not the cause of my PD. I therefore was not included in the LRRK-2 research cohort, but its findings may very well help the vast majority of PD patients who do not carry that mutation. I participated for several years in other online research projects conducted by 23andMe, which helped to define the natural history of PD and correlate it with the findings in their large genetic database. As a result of their research, a number of new mutations have been discovered to be associated with PD (1). The FDA’s action has shut down this valuable research effort, along with the direct-to-consumer genetic testing service.

As a PD patient, I discussed my 23andMe genetic profile with my treating neurologist, and I requested that he include it on my chart. My main concerns as a patient are:

1) Are my genetic test results accurate, as reported by 23andMe ?

2) Are the reported statistical associations with diseases accurate ?

It is appropriate for the FDA to monitor the accuracy of test results and associated interpretive data supplied by direct-to-consumer genetic test providers. However, the FDA should not hold test providers responsible when consumers misuse this data for unintended purposes, such as for adjusting their warfarin dose without INR testing, or as a substitute for recommended cancer screening tests or regular medical care.

To address the clinical scenario posed in the editorial, here is what I would do if an asymptomatic patient came to my office with a direct-to-consumer genetic test positive for a mutation conferring elevated risk for Crohn’s disease. First, I would reassure him that having a genetic variation associated with increased risk of a disease does not mean that disease will necessarily develop, and many people exhibit such variations and live long and healthy lives. I would add that Crohn’s disease, like other immune disorders, exhibits substantial discordance between identical twins, and thus must have significant environmental risk factors (2). Next, I would perform a complete history and physical exam, appropriate for his age and condition and concerns, including careful examinations of his mouth, abdomen, and anus. I would make certain that he was up to date on his colon cancer screening, per the current guidelines. I would discuss the signs and symptoms of Crohn’s disease with him, and give him a stool test kit for occult blood and draw basic blood labs to check for occult anemia and iron and B-12 deficiency, and any basic labs he might be due for. I would tell him that smoking seems to worsen Crohn’s disease, and offer him assistance with smoking cessation. I would ask him to schedule a follow-up visit in a few weeks, and to call me in the unlikely event he developed any of the signs or symptoms of Crohn’s disease. During the intervening weeks, I would study his genetic report and phone a GI colleague for advice. If the patient is at elevated risk of Crohn’s disease, an early diagnosis would allow time for smoking cessation and surveillance to reduce the risk of future fistulas and other complications.

1: 23andMe Parkinson’s disease research results website, accessed on 2/12/2014: http://blog.23andme.com/23andme-research/23andme-and-parkinsons-past-present-and-future/

2: Orholm M, Binder V, Sørensen TI, Rasmussen LP, Kyvik KO. Concordance of inflammatory bowel disease among Danish twins. Results of a nationwide study. Scand J Gastroenterol. 2000 Oct;35(10):1075-81. PubMed PMID: 11099061

On 2014 Mar 31, Rebecca Balter commented:

The findings from my article are misrepresented in the introduction

The sentence from the article says: "In addition, a withdrawal from continuous administration of morphine attenuates the increase in thermal sensitivity seen in morphine -treated mice [21]."

This should say: In addition, access to a running wheel can attenuate the increase in thermal sensitivity seen during withdrawal from continuous administration of morphine.

On 2014 Sep 09, Joseph Antony commented:

Great Paper! However, does not mention our work in successfully infecting macaques with human VZV (Willer et al.,2012)

On 2014 Feb 27, FREDERICK DOMANN commented:

These interesting findings support and extend prior work by the same group (PMID: 16170370) and add pancreatic cancer to the growing list of malignancies where alterations in EcSOD expression can affect malignant phenotype (see also PMIDs: 23318435, 22064654, and 19602586). Nevertheless, Figure 7D unfortunately omits an important node in the signaling pathway between superoxide and HIF-1a; that is the effector molecule, the HIF-1 prolyl hydroxylases 1-3 (PHD1-3) which are the direct targets inhibited by superoxide (www.dovepress.com/getfile.php?fileID=17843). Conversely, and logically then, elevated SOD activity inhibits the inactivation of PHD family enzymes and suppresses the hypoxic induction of HIF-1a.

On 2014 Mar 09, Gyanshankar Mishra commented:

We have also done a prior study on Tuberculosis Prescription Practices In Private And Public Sector In India: Mishra G, Mulani J. Tuberculosis Prescription Practices In Private And Public Sector In India. NJIRM. 2013; 4(2): 71-78.

Available online at http://www.scopemed.org/?mno=36915

On 2014 Mar 26, Tom Kindlon commented:

Various responses to this paper have been posted here: http://bmjopen.bmj.com/content/4/2/e003973/reply including two by me: (i) "High rates of deterioration following graded exercise therapy and cognitive behavioural therapy have been reported in patient surveys" http://bmjopen.bmj.com/content/4/2/e003973/reply#bmjopen_el_7699 and (ii) "Re:Re:High rates of deterioration following graded exercise therapy and cognitive behavioural therapy have been reported in patient surveys" http://bmjopen.bmj.com/content/4/2/e003973/reply#bmjopen_el_7774

On 2014 Feb 11, Ellen M Goudsmit commented:

Brurberg et al's analysis of the various criteria for ME/CFS is timely and important[1]. However, the information about the 'London' criteria (LC) for classic ME is misleading and they missed the more recent, updated case definition [2]. The LC devised by Dowsett et al were not published in the Westcare Taskforce Report. The latter reproduced a rewritten version by an unknown person which missed the last part of the paper. To my knowledge, that version was never used. In contrast, the LC were used in at least four studies, sometimes alongside the Oxford criteria. I know that as I was the Chair of the Research Working Group at AFME at the time and involved in the funding of studies. We had few conditions but one was the use of the LC. Consequently, AFME was probably the only organisations to continue to study classic ME after the introduction of the CDC and Oxford criteria when attention had been diverted to CFS. Some of the published papers refer to the LC in the text and references; others to criteria developed by AFME, or wrongly, to the TaskForce report. With only one exception, all the studies using the LC selected a homogeneous group with abnormalities in 100% of those tested. Of these, Paul et al (1999) revealed that it was possible to objectively measure the cardinal symptom of classic ME (LC: criterion 1). This information is included in the revised guidelines published in 2009 [2].

Until two years ago, journals had little interest in classic ME, that is, the illness described by physicians since the 1950s. Journals rejected revised guidelines, not because they were poor but because the whole area of criteria was deemed too ‘contentious’. This undermined the scientific process as those reviewing other criteria were not aware of knowledge obtained by colleagues. The new case definition for classic ME was eventually accepted by the Health Psychology Update (British Psychological Society), but the update is only available online [3].

It may well be that there is no difference between samples selected using the CDC, and the newer criteria for ME/CFS and classic ME. However, assumptions require testing which is why scientists still require sound case definitions for ME. They remain an important resource for doctors and researchers wishing to increase diagnostic precision. Conversely, editorial policies that reject proposed criteria for political reasons are unhelpful and will only result in incomplete assessments and analyses.

1. BMJ Open 2014 4:e003973; doi:10.1136/bmjopen-2013-003973

2. Goudsmit EM, Shepherd C., Dancey CP, Howes S. ME: Chronic fatigue syndrome or a distinct clinical entity? Health Psychology Update, 2009;18(1):26-33. http://www.bpsshop.org.uk/Health-Psychology-Update-Vol-18-No-1-2009-P797.aspx Updated by EMG in 2012 and available from: http://www.foodsmatter.com/me_and_cfs/cfs_me_causes_general/articles/goudsmit-me-clinical entity-10-12.html

3. Howes S, Goudsmit E, Shepherd C. Myalgic encephalomyelitis (ME). Criteria and clinical guidelines 2014. Available from: http://www.axfordsabode.org.uk/me/mecrit2014.htm

On 2015 Aug 13, Lydia Maniatis commented:

The authors' say that "The data also argue against an image decomposition mechanism" as an explanation of the snake illusion. This claim hinges on having controlled for apparent transparency/illumination effects in some of their stimuli. However, as they themselves acknowledge in the article, their stimuli still present such effects, despite the investigators' having removed some features, like X-junctions, commonly associated with transparency effects. Thus, their claim vis a vis image decomposition cannot be said to have been corroborated.

On 2014 Mar 29, Tero Kivelä commented:

Thank you very much for useful clarifications!

On 2014 Mar 24, Robert M Verdijk commented:

The current TNM classification was published when this study was ongoing. The aim of this study was to evaluate the histopathologic confirmed presence of extraocular extension and therefore we only included enucleated eyes in our study. Determination of the size of the extraocular extension after eye-conserving therapy is not completely independent from interpretation, and every other type of examination than histopathologic examination will not be as specific to determine microscopic extraocular extension. In our multivariate analysis a larger episcleral diameter of extraocular extension (HR=1.078) and presence of chromosome 8q gain (HR= 2.874) were correlated with worse survival. As chromosome 8q gain is associated with a larger tumor size (van den Bosch et al., 2013) we added tumor prominence and tumor size in our model in order to diminish the effect of these parameters. To answer the questions from Kivelä we also analyzed the tumors enucleated before 1999 and after 1999 separately, and found no differences in survival. The parameter that was strongly associated with survival before and after 1999 was gain of chromosome 8q, consistent with what we already reported, nevertheless, indeed we agree with Kivelä as discussed in our paper that one should be careful in extrapolation of the data.

In addition we did not evaluate macrophage infiltration separately in this study, and no immunohistochemistry was used. Extracellular matrix patterns were evaluated within this study, but no immunohistochemistry for microvessel density measurement was performed. We did not choose to evaluate these parameters. For extracellular matrix patterns closed loop networks were evaluated and defined as at least three back to back loops. Furthermore these were evaluated using non-counterstained PAS stain. A dark green filter was not used since in our experience closed loop networks can be identified without the filter. In all our studies closed loop networks correlated with prognosis.

We reported in our paper that two patients treated with fractionated stereotactic radiotherapy had extraocular extension. For both patients the B-scan did not show evident extraocular extension at time of diagnosis. After enucleation one patient had an extraocular extension of 0.10 mm, and for the other patient the extension was 3.0 mm. The indication for enucleation in these patients was secondary glaucoma and intraocular tumor progression, respectively.

As requested by Kivelä we constructed Kaplan-Meier curves according to the separate groups for extraocular extension, as subdivided in the 7th edition of the TNM classification for uveal melanoma. The survival curves and show a decreased survival in larger episcleral diameter of extraocular extension. These supplementary Figures 1 and 2 could not be introduced in this reply, but can be requested by email and will be available through my Researchgate account.

References: van den Bosch T, van Beek JG, Vaarwater J, Verdijk RM, Naus NC, Paridaens D, de Klein A, Kiliç E. Higher percentage of FISH-determined monosomy 3 and 8q amplification in uveal melanoma cells relate to poor patient prognosis. Invest Ophthalmol Vis Sci. 2012 May 14;53(6):2668-74.

On 2014 Mar 11, Tero Kivelä commented:

This is an important contribution, because it is the first study that lends independent support to the new Tumor, Node, Metastasis (TNM) classification subcategories for extraocular extension of uveal melanoma that were introduced in the 7th edition of this classification (Kujala E, 2013). The new size categories already had been independently confirmed (Shields CL, 2013).

The subdivisions for extraocular extension in TNM were based on a smaller data set than the size categories, and the independent significance of extraocular extension moreover had been challenged (Coupland SE, 2008) because another dataset suggested that extraocular extension reflected tumor malignancy and would not be significant after adjusting for tumor size, presence of epithelioid cells, closed extravascular matrix loops, high mitotic rate, and monosomy 3. The TNM bulding data set did not contain histopathologic and genetic variables, and could not address this criticism, whereas the present study adjusted for all these factors and still found extraocular extension, especially its size, to be independently associated with metastatic death. This is encouraging but not yet final proof.

All studies have some limitations and the following will affect interpretation of the present one:

1: The material was enriched in large tumors: it was unselected between 1987 and 1999, but between 1999 and 2011 only large melanomas (diameter >16 mm and thickness >12 mm) were enucleated and thus were available for analysis. The reported statistics thus do not directly apply to consecutive, unselected uveal melanomas (because statistics can only be extrapolated to a population that is similar to the sample).

2: Inflammation was roughly determined by presence of obvious clusters of lymphoid inflammatory cells, whereas macrophage infiltration also is associated with survival and with monosomy 3 in uveal melanomas (e.g. Mäkitie T, 2001, Maat W, 2008, Bronkhorst IH, 2011).

3: Microvascular density was not analyzed (e.g. Mäkitie T, 1999, Chen X, 2002); it is associated with prognosis of uveal melanoma independent of cell type, extravascular matrix patterns and inflammation (macrophages) and might have entered the model instead of another variable.

4: Kaplan-Meier graph by the diameter of the extraocular extension, which would have allowed direct comparison with the TNM data, is absent.

5: The paper does not explicitly mention:

5.1: Whether secondarily enucleated eyes had an extraocular extension already at the time of radiotherapy or developed it later.

5.2: What extracellular matrix patterns refer to in this study. They were coded as being absent or present. However, all uveal melanomas have at least one of the nine described patterns (even absence of vessels is a pattern, i.e. silent; Folberg R, 1993).

5.3: Were extracellular matrix patterns identified from non-counterstained sections under a dark green filter as originally described (Folberg R, 1993) or from counterstained sections in which they may be less obvious (McLean IW, 1997).

Conflict of interest: I was one author of the uveal melanoma chapter of the 7th edition of the TNM classicifation, American Joint Committee on Cancer.

On 2015 May 17, Tim Smits commented:

Below is a Letter submitted to the Lancet concerning this publication. Lancet did not accept this Letter, though it addresses a critical aspect of the preregistration process (well, actually it was postregistration). For a visual timeline of the errors in the preregistration process, I refer to a blogpost http://persuasivemark.blogspot.be/2014/03/the-pitfalls-of-pre-registration.html. Such major errors in the preregistration do make the benefits of such quality control measures obsolete and could provide a cover-up for researcher degrees of freedom.

In their article on non‐medicated cognitive therapy for schizophrenia, Morrison and colleagues[1] report on significant reductions in psychiatric symptoms. These interpretations are rash, due to two severe limitations. The protocol for the study is difficult to find and reconstruct. The lead author registered[2] the trial on October 21st 2010, about eight months after data collection started, without making reference to planned analyses. An article with a more extensive protocol was published[3] in 2013. It was first submitted in October 2012, thus compromising the 9 to 18‐month study duration ending in February 2013. That article did, however, include more detailed analysis plans: “…analysis of repeated measures using a mixed‐effects model…”. Although the proposed testing of treatment effects at the individual level seems the best approach, the final Lancet article only reports effects between treatment and control groups at different time points. Clearly, this is an inferior analysis of a longitudinal repeated‐measures design, and it violates the aforementioned protocol[3] . The published analyses fail to take into account interpersonal variability at the onset of data collection. They also fail to capitalize on the design’s capability to identify individual treatment effects.<br> Ironically, Morrison and colleagues claim to report on all outcomes specified in their protocol, but they fail to report on their planned analyses. Such conduct increases researcher degrees of freedom, while simultaneously obscuring the use of this freedom with the protocol registration allegedly backing up the reported research practices. As others have claimed previously[4,5], the implications of such practices are potentially serious.

References

1 Morrison AP, Turkington D, Pyle M, Spencer H, Brabban A, Dunn G, Christodoulides T, Dudley R, Chapman N, Callcott P, Grace T, Lumley V, Drage L, Tully S, Irving K, Cummings A, Byrne R, Davies LM, Hutton P. Cognitive therapy for people with schizofrenia spectrum disorders not taking antipsychotic drugs: a single‐blind randomised controlled trial. Lancet 2014 ahead of print

2 http://www.controlled‐trials.com/ISRCTN29607432/morrison

3 Morrison AP, Wardle M, Hutton P, Davies L, Dunn G, Brabban A, Byrne R, Drage L, Spencer H, Turkington D. Assessing Cognitive Therapy Instead Of Neuroleptics: Rationale, study design and sample characteristics of the ACTION trial. Psychosis 2013; 5(1): 82‐92.

4 Schulz KF, Altman DG, Moher D. Protocols, probity, and publication. Lancet 2009; 373: 1524.

5 Glasziou P, Altman DG, Bossuyt P, Boutron I, Clarke M, Julious S, Michie S, Moher D, Wager E. reducing waste from incomplete or unusable reports of biomedical research. Lancet 2014; 383: 267‐76.

On 2014 Apr 17, James C Coyne commented:

This abstract is exceedingly misleading for a study that was registered after enrollment started and without adequate designation of which time point which outcome would be assessed as primary. At the end of the intervention, there were no significant differences between CBTp and treatment as usual.

In 2 blog posts at PLOS Mind the Brain, I discuss the serious problems with this study:

http://blogs.plos.org/mindthebrain/2014/02/25/much-ado-little-lancet-study-cognitive-therapy-persons-unmedicated-schizophrenia/

http://blogs.plos.org/mindthebrain/2014/03/11/much-ado-modest-misrepresented-trial-cbt-schizophrenia-part-2/

The trial really did not producing usable data concerning the efficacy of cognitive behavior therapy for patients with unmedicated schizophrenia because of

An unusually mixed group of patients participating in the study.
An inappropriately constructed control group that does not represent conditions in routine care nor as a composite allow meaningful comparisons with the active intervention, CBTp.
Substantial loss to follow-up from an already small exploratory study.
A decision of the investigator team to abort long-term follow-up but proceed with data analysis as if this decision had not been made.
A substantial number of patients in both the intervention and control group receiving antipsychotic medication, including those in the control group who showed the greatest improvement.

I could go on but best to see these critisms and others elaborated with others at my blog posts.

On 2014 Aug 16, Raha Pazoki commented:

“Getting the Most out of Available Resources”

One of the biggest challenges in the world of genomics is reaching enough sample size in order to identify extremely small effects of genetic loci on risk of complex diseases. Large cohort studies in developing world such as the Persian Gulf Healthy Heart Study are precious resources in this regard. While the research budget in developing world is limited, allocation of international financial support may help these cohort studies contribute to better understanding of gene-disease associations.

Conflicts of interest: I've worked with the Persian Gulf Healthy Heart Study in the past.

On 2014 Apr 12, John Sotos commented:

There is an interesting "control" study of an unfortunate patient who presented with a similar cardiac syndrome, but whose physician was not steeped in Dr. House.

http://www.ncbi.nlm.nih.gov/pubmed/24499215

On 2014 Feb 12, Andrea Messori commented:

Co-authors of this Note: Valeria Fadda, Roberta Gatto, Dario Maratea, and Sabrina Trippoli (all from the HTA Unit of Estav in Firenze, Italy)

Relative risk (RR) and risk difference (RD) have different advantages and disadvantages [see Walter (2000) for further discussion]. In the paper by Stidham and co-workers (2014), the analysis focused on the induction of remission is the one endowed with the main therapeutic implications. To better explore this data set (end-point = induction of remission), we have re-analyzed the results from the various trials by using both RR and RD as outcome measures for the meta-analysis. Our results are shown in a figure that can be downloaded from the following link: www.osservatorioinnovazione.net/papers/stidham-reanalysis.pdf .

The analysis based on RDs is advantageous because the outcome measure is an absolute one and permits us to interpret the clinical significance of these findings using the number need to treat.

FIGURE LEGEND: The meta-analysis shown in this figure re-examined the same information previously reported by Stidham et al. (data set: induction of remission). Our figure shows the Forest plot with the values of RR (Panel A) or RD (Panel B) for individual trials (solid square with 95%CIs indicated by horizontal bars) and for some trial subgroups (diamonds in yellow). The pooled rates for the entire data-set are shown as blue diamonds. I² is a measure of heterogeneity. Statistical calculations were performed by the OMA software (Open Meta-Analyst version 4.16.12, Tufts University, U.S., url http://tuftscaes.org/open_meta/). Abbreviations: Ev, number of events; Trt, number of patients receiving treatment.

REFERENCES:

-Stidham RW, Lee TC, Higgins PD, Deshpande AR, Sussman DA, Singal AG, Elmunzer BJ, Saini SD, Vijan S, Waljee AK. Systematic review with network meta-analysis: the efficacy of anti-tumour necrosis factor-alpha agents for the treatment of ulcerative colitis. Aliment Pharmacol Ther. 2014 Feb 9. doi: 10.1111/apt.12644. [Epub ahead of print]

-Walter SD. Choice of effect measure for epidemiological data. J Clin Epidemiol. 2000 Sep; 53(9):931-9.

On 2014 Feb 12, Hilda Bastian commented:

It would be wonderful if as many post-stroke therapies were as effective, and the evidence for them as strong, as this review concludes. Unfortunately, that's not the case.

The abstract of this review talks about trials in over 25,000 patients - but it doesn't point out that the numbers for individual interventions is, with only some exceptions, small. The review has several major flaws, in particular having no protocol to guard against problems caused by multiple testing and subgroup analyses. Crossover trials are pooled with parallel trials, and the effect of this on the various analyses is not clear: methodological characteristics of the individual trials are not reported. A scoring method is used for the individual trials, for which only the summary score is available.

In addition, it's important to note that the search for this review was done in June of 2011. As well as using more robust methods, other reviews are significantly more up-to-date, e.g. systematic reviews on treadmills (Mehrholz J, 2014) and physical fitness training (Saunders DH, 2013).

Although this review's abstract and conclusions are strongly positive about 30 interventions they consider, the authors do point out in the discussion that: "well controlled, dose-matched trials with significant effects in favor of the experimental intervention have been rather scarce."

For a good overview to consider alongside well-conducted recent systematic reviews, see Langhorne P, 2011.

On 2016 Aug 30, Ben Goldacre commented:

This trial has the wrong trial registry link associated with it on PubMed: both in the XML on PubMed, and in the originating journal article. The ID in the link provided is NCT016301952912. We believe the correct link, as found elsewhere in the text, is NCT01952912.

This comment is being posted as part of the OpenTrials.net project^[1] , an open database threading together all publicly accessible documents and data on each trial, globally. In the course of creating the database, and matching documents and data sources about trials from different locations, we have identified various anomalies in datasets such as PubMed, and in published papers. Alongside documenting the prevalence of problems, we are also attempting to correct these errors and anomalies wherever possible, by feeding back to the originators. We have corrected this data in the OpenTrials.net database; we hope that this trial’s text and metadata can also be corrected at source, in PubMed and in the accompanying paper.

Many thanks,

Jessica Fleminger, Ben Goldacre*

[1] Goldacre, B., Gray, J., 2016. OpenTrials: towards a collaborative open database of all available information on all clinical trials. Trials 17. doi:10.1186/s13063-016-1290-8 PMID: 27056367

* Dr Ben Goldacre BA MA MSc MBBS MRCPsych<br> Senior Clinical Research Fellow<br> ben.goldacre@phc.ox.ac.uk<br> www.ebmDataLab.net<br> Centre for Evidence Based Medicine<br> Department of Primary Care Health Sciences<br> University of Oxford<br> Radcliffe Observatory Quarter<br> Woodstock Road<br> Oxford OX2 6GG

On 2015 Jan 28, David Keller commented:

See also a related comment on PubMed Commons at the following link

http://www.ncbi.nlm.nih.gov/pubmed/25264620#cm25264620_6445

On 2015 Dec 03, Siddharudha Shivalli commented:

I read the article titled ‘Smear positive pulmonary tuberculosis among diabetic patients at the Dessie referral hospital, Northeast Ethiopia’ by Amare H et al, with a great interest. Authors’ efforts are praiseworthy. In their single centre hospital based study, authors highlight the prevalence of TB among known diabetics and factors associated with it. However, following are some issues and concerns.

For cross sectional study, adequacy and representativeness of study sample size are essential to ensure the validity of the study findings. Authors have justified the adequacy by calculating the sample size (n=236), however, I am not sure about the representativeness. Do 236 study participants selected consecutively over a period of only 3 months (February 2012 to April 2012) represent the diabetic patients who visit the Dessie referral hospital (average diabetic patient number approximately 1,700)? Systematic random sampling would have been more apt for this i.e. including every 4th or 5th eligible patient depending upon the weekly or monthly patient input.

Another limitation of the study is the inclusion criterion as authors have studied only pulmonary tuberculosis (PTB) suspected diabetic patients. If one wants to estimate the prevalence, all the diabetics should have been studied. Hence, reported prevalence in this study may be an underestimation. In addition reporting of prevalence should have been done with 95% confidence intervals (6.2%, 95% CI: 3.7-10.25). In addition, reported associations between prevalence of PTB among diabetes patients and study variables in this study may not imply cauasality owing to cross sectional study design.

In this study, variables which had a p-value of less than 0.20 were taken to multivariate logistic regression. However, it is recommended to assess and report the adequacy of applied regression model. Failure to do so may lead to misleading or incorrect deductions. Although the study sample was relatively large (n=236), a word about R2 (explaining the variance in prevalence of PTB) of the applied regression model would have been more affirmative.

None the less, I must congratulate the authors for investigating an important public health problem.

Competing interests: The author declares that there is no conflict of interest about this publication.

On 2014 Apr 12, John Sotos commented:

Allen et al (1) courageously report a woman who underwent heart transplantation when her cardiomyopathy’s reversible cause – arthroprosthetic cobaltism (APC) from bilateral metal-on-metal hips – went undiagnosed. Endorsing their conclusion that clinicians in cardiac, orthopedic, thyroid, rheumatic, and ophthalmic specialties need improved awareness of this multi-system disorder, we would add neurologists, psychiatrists, and, especially, primary care physicians.

Cobalt causes a full spectrum of neuropsychiatric effects, from anxiety and irritability to life- threatening mood and thought disorders, plus peripheral neuropathy, cranial neuropathy, cognitive decline, and gait disorders (2,3).

Primary care physicians are likely to encounter APC early in its course, when its manifestations – including tinnitus, fatigue, disturbed sleep, nausea, “mental fog,” and headaches – are mild, non-specific, and easily dismissed as simple aging (4,5).

However, because APC is both progressive and reversible, we suggest all physicians adopt a low threshold for checking cobalt levels in at-risk patients, even those without hip complaints and those with metal-on-plastic or metal-on-ceramic hips (3).

(1) Allen LA, Ambardekar AV, Devaraj KM, Maleszewski JJ, Wolfel EE. Missing elements of the history. N Engl J Med. 2014; 370: 559-566.

(2) Sotos JG, Tower SS. Systemic disease after hip replacement: aeromedical implications of arthroprosthetic cobaltism. Aviation, Space, and Environmental Medicine 2013; 84: 242-245.

(3) Catalani S, Rizzetti MC, Padovani A, Apostoli P. Neurotoxicity of cobalt. Hum Exp Toxicol. 2012; 31: 421-437.

(4) Tower SS. Arthroprosthetic cobaltism: neurological and cardiac manifestations in two patients with metal-on-metal arthroplasty: a case report. J Bone Joint Surg Am. 2010; 92: 2847-2851.

(5) Leikin JB, Karydes HC, Whiteley PM, Wills BK, Cumpston KL, Jacobs JJ. Outpatient toxicology clinic experience of patients with hip implants. Clin Toxicol (Phila). 2013; 51: 230-236.

On 2014 Mar 09, David Keller commented:

In this case of a 67-year-old woman nursing home resident with fever, tachypnea, confusion, uremia and bilateral pneumonia, I question the decision not to obtain blood cultures, and to treat blindly with empiric therapy for community-acquired pneumonia: ceftriaxone and azithromycin. Some of her close contacts among the staff and residents of her nursing home are likely to be colonized with Pseudomonas and other resistant gram negatives. In this nursing home patient severely ill with bilateral pneumonia being considered for ICU admission, I would start with broad coverage of resistant pathogens, obtain blood cultures, and narrow the therapeutic spectrum as soon as warranted by the patient’s clinical status and culture results. I agree with the need to limit the use of broad-spectrum antibiotics, but it seems like too much of a gamble to leave gaps in her coverage, without drawing cultures, when she is this ill.

1: Wunderink RG, Waterer GW. Clinical practice. Community-acquired pneumonia. N Engl J Med. 2014 Feb 6;370(6):543-51. doi: 10.1056/NEJMcp1214869. PubMed PMID: 24499212.

On 2014 Feb 15, Amanda Capes-Davis commented:

Cell line names can be very confusing and there are a few issues here to be aware of. KB is not epidermoid carcinoma, as originally thought, but is actually cross-contaminated with HeLa, a human cervical adenocarcinoma cell line. LU-1 is described here as lung adenocarcinoma, but there are two similarly named cell lines in the literature - SK-LU-1 and LU. SK-LU-1 is known be authentic, while LU is cross-contaminated with HeLa. Which did the authors use?

For a list of known cross-contaminated cell lines, see http://iclac.org/databases/cross-contaminations/.

On 2017 Aug 21, John M Darlow commented:

Dear Anne, Thank you for detecting that there is a small error, but you have corrected the wrong bit! As you can verify by typing the rs numbers into, e.g., the UCSC Genome Browser, the two SNPs in question ARE in the gene KIAA1324, on 1p13.3, as correctly stated in the paper. The error is that the old name of the gene was 'EIG121' not 'EIG121L'. If you had looked up the gene KIAA1324L (old name EIG121L) you would have found that that gene is on 7q21.12!

On 2017 Aug 15, Anne Niknejad commented:

Error:

'There are two adjacent significant results on 1p13.3 in KIAA1324 alias EIG121L,...'

should be

'There are two adjacent significant results on 1p13.3 in KIAA1324L alias EIG121L,...'

On 2014 Nov 22, Ruchira Engel commented:

We reviewed this article in our journal club. Our comments can be found on our blog Sanquin Digests.

On 2014 May 11, Dorothy V M Bishop commented:

For a critique and list of concerns about RDoC, please see http://deevybee.blogspot.co.uk/2014/05/changing-landscape-of-psychiatric.html

On 2014 Mar 10, Gaetano Santulli commented:

Dr. Marrouche and colleagues (1) found in their elegant study that left atrial fibrosis, quantified by delayed enhancement magnetic resonance imaging (DE-MRI) is independently associated with likelihood of recurrent arrhythmia in patients with atrial fibrillation (AF) undergoing catheter ablation. Their results also reveal that a hypertensive state was significantly associated with the amount of atrial fibrosis. The potential explanations reported by the Authors to discuss the relationship between hypertension and atrial fibrosis appear not completely satisfactory. Indeed, the Authors considered the hypertensive disease just as a discreet, not continue, variable, defined as systolic blood pressure > 160 mmHg, without providing any information on the pharmacological regimen of the enrolled patients. Given the acknowledged functional role of specific anti-hypertensive drugs, including angiotensin converting enzyme inhibitors and angiotensin receptor blockers, in preventing atrial electrical and structural remodelling (2), it would be of interest to see the results of their analysis conducted considering these parameters. It also would be interesting to know the influence of statins or polyunsaturated fatty acids (3), since the Authors show that 30% of their patients had dyslipidemia. Lastly, several studies demonstrated that patients with AF display a reverse atrial remodelling at 1-year follow up after ablation, evaluated via ultrasound analysis or through inflammatory markers, collagen turnover, and natriuretic peptides (4). Do the Authors have any data on atrial remodelling?

Conflict of Interest Disclosures: None.

References 1. Marrouche NF, Wilber D, Hindricks G, et al. Association of atrial tissue fibrosis identified by delayed enhancement MRI and atrial fibrillation catheter ablation: the DECAAF study. JAMA. Feb 5 2014;311(5):498-506. 2. Ehrlich JR, Hohnloser SH, Nattel S. Role of angiotensin system and effects of its inhibition in atrial fibrillation: clinical and experimental evidence. European heart journal. Mar 2006;27(5):512-518. 3. Savelieva I, Camm J. Statins and polyunsaturated fatty acids for treatment of atrial fibrillation. Nature clinical practice. Cardiovascular medicine. Jan 2008;5(1):30-41. 4. Reant P, Lafitte S, Jais P, et al. Reverse remodeling of the left cardiac chambers after catheter ablation after 1 year in a series of patients with isolated atrial fibrillation. Circulation. Nov 8 2005;112(19):2896-2903.

Celestino Sardu, MD (¹), Gaetano Santulli, MD, PhD,(²) ¹Second University of Naples, Naples, Italy; ²Columbia University, New York, NY, USA

On 2016 Aug 24, Ben Goldacre commented:

This trial has the wrong trial registry ID associated with it on PubMed: both in the XML on PubMed, and in the originating journal article. The ID given is NCT00123456. We believe the correct ID, which we have found by hand searching, is NCT00675324.

This comment is being posted as part of the OpenTrials.net project^[1] , an open database threading together all publicly accessible documents and data on each trial, globally. In the course of creating the database, and matching documents and data sources about trials from different locations, we have identified various anomalies in datasets such as PubMed, and in published papers. Alongside documenting the prevalence of problems, we are also attempting to correct these errors and anomalies wherever possible, by feeding back to the originators. We have corrected this data in the OpenTrials.net database; we hope that this trial’s text and metadata can also be corrected at source, in PubMed and in the accompanying paper.

Many thanks,

Jessica Fleminger, Ben Goldacre*

[1] Goldacre, B., Gray, J., 2016. OpenTrials: towards a collaborative open database of all available information on all clinical trials. Trials 17. doi:10.1186/s13063-016-1290-8 PMID: 27056367

* Dr Ben Goldacre BA MA MSc MBBS MRCPsych<br> Senior Clinical Research Fellow<br> ben.goldacre@phc.ox.ac.uk<br> www.ebmDataLab.net<br> Centre for Evidence Based Medicine<br> Department of Primary Care Health Sciences<br> University of Oxford<br> Radcliffe Observatory Quarter<br> Woodstock Road<br> Oxford OX2 6GG

On 2014 Jun 05, Robert M Flight commented:

Unless I misread the paper, the NMAD that is used for assigning weights is derived from the exact same data that subsequently undergoes biclustering. A better approach might be to derive weights from independent experiments and apply those to the data that one wishes to bicluster.

In addition, the difference in significance with and without (alpha=0) the incorporation of the weights seems to imply that the use of the weights adds very little information to the biclustering algorithm, and makes one question as to why they should be calculated at all. The biclustering algorithm using the networks shows a definite difference compared to the other methods tested, but the weighted vs unweighted biclustering shows very little differences, making one question the significance of the "weighted co-clustering approach", vs a "network based co-clustering approach".

On 2017 Jul 28, Randi Pechacek commented:

Jonathan Eisen mentioned this article on a microBEnet blog post while discussing the study of microbiomes of the built environment.

On 2014 Mar 14, Jonathan Eisen commented:

Brooke Borel has an article in Popular Science about this paper. See Building Design Influences Bacterial Growth

On 2016 Jan 15, CREBP Journal Club commented:

The presented article is an interesting example of high quality well-conducted overview of systematic reviews. The authors concluded that EBHC teaching strategies should focus on implementing multifaceted, clinically integrated approaches with assessment. Our journal club discussed the minimum components for EBHC intervention that could be equally effective, and the equivalence between lecture-based and online EBHC training which resonate the findings of a recent RCT of blended learning vs. didactic learning approaches for teaching EBHC (Ilic D, 2015). We have also discussed the inconsistencies in describing the content of EBHC educational interventions in the included separate studies which impede the replication and implementation of their findings. We referred to the currently developing reporting guideline for educational intervention for EBP (Phillips AC, 2014).<br> Our Journal club have also discussed the heterogeneity of outcome measures both between and within included systematic reviews which prevent the authors from providing a pooled effect estimate of the effect of teaching EBHC (Shaneyfelt T, 2006). It is worthwhile to have acceptable standardised outcome measures to assess the effect of teaching EBHC as suggested by Sicily statement (Tilson JK, 2011).

See CREBP Journal Club for more information.

On 2014 Mar 06, nagendra singh chauhan commented:

very nice article

On 2014 Feb 17, Jesus Mendez-Gonzalez commented:

This is a very interesting study in terms of early diagnosis of lung cancer. DNA methylation markers are especially promising in this area, as they can be studied in multiple specimens, arise even in premalignant lesions and could be used to complement low-dose CT screening, reducing its associated and problematic high false positive rate. To date, most studies have focused on hypothesis-driven targets, thus reducing the chance of identifying the best (but hidden) candidates. However, and importantly, the authors chose a wide, blind and functional identification method and validated the best candidates taking advantage of the TCGA database and two additional cohorts of lung cancer samples. They, finally, come up to three methylation markers that show an impressive sensitivity with 100% specificity. The real value of this panel is still to be determined: no data in fluid samples (e.g. bronchoaspirates or sputum) is available, and MSP technique (which has to deal with the problem of false positives due to potential incomplete bisulphite conversion) was only performed in 7 normal samples. However, the approach is really encouraging. The potential prognostic value of these alterations was also explored, but no relation to survival was found. For the authors this is somehow expected, due to the absence of “an established role in the pathogenesis of lung cancer and/or an extremely high prevalence of methylation”. This is reasonable, but there are some points that would be worth to discuss:

• As the authors point, “TCGA samples are not annotated for therapies received, therefore no control for treatment in analysis is possible”. Surely, recurrence after surgery –especially in early-stages tumors, where no adjuvant therapy is usually administered- would be a better end point, but no information is available.

• Unfortunately, this paper was published shortly after this one: “A prognostic DNA methylation signature for stage I non-small-cell lung cancer” ( http://www.ncbi.nlm.nih.gov/pubmed/24081945 ). Thus, there was no possibility of discussing the commonalities and differences that both studies found. In this article we analyzed the DNA methylation of a wide amount of NSCLC samples through the Infinium 450k array, the same platform as TCGA used (these data are also available for public analysis). Consistently, CDO1 and HOXA9 were found as largely differentially methylated between tumors and normal tissues. Additionaly, we studied the relationship between DNA methylation and recurrence in stage I resected tumors, where no adjuvant and potentially confounding treatments were given. Contrary to Wrangle et al. we found that HOXA9 promoter hypermethylation correlated with a worse progression free survival. These data were validated by pyrosequencing in an independent cohort.

  Two main differences have to be underscored between the studies. The first one has to do with the different end-points analyzed (time to death vs progression free survival). The second one is that we set a higher threshold to classify a sample as hypermethylated (0.4 vs 0.2). If our results are validated in larger and independent cohorts, there are at least two possibilities to explain these results: i) HOXA9 exerts an unknown role in the pathogenesis of lung cancer (some small studies suggest this option: http://www.ncbi.nlm.nih.gov/pubmed/21757291) and ii) HOXA9 “heavy” hypermethylation is a marker of particular malignant lesions more prone to show an aggressive behavior.

On 2014 Feb 04, Jan H. Duedal Rölfing commented:

Free full-text available from: http://informahealthcare.com/doi/full/10.3109/17453674.2013.869716

On 2014 Apr 26, Gonzalo Sanchez commented:

In order to explain concurrent signs of spastic hemiparesis with bleeding from the nose and the ear in the closed head injury of Case # 8 of the Edwin Smith Papyrus, JC Ganz states that recent trauma must have occurred in a patient with an already existing hemiparesis. This would be a good explanation if the Papyrus were describing a specific injury in a specific patient. The Edwin Smith papyrus is, rather, a teaching trauma treatise of “Case Types” with Case #8 addressing Closed Head Injuries. Sanchez and Meltzer (2012)1 note (p.5) their clinical interpretation is based on the textual evidence and the structure of the original document.<br> In Appendix II these authors acknowledge Case #8a as a closed head injury that has passed the acute stage “ as development of spasticity takes several weeks”. Fresh bleeding through the nose and the ears would indeed be unlikely present at this stage. Apparent inconsistencies in these clinical issues may be simply related to the various findings observed by the ancient Egyptians in cases of the same type. It is our opinion that strict criticism of the ancient physicians’ clinical methodology in structuring and documenting their teaching text cannot be applied using current criteria.

Gonzalo M. Sanchez MD. and Edmund S. Meltzer Ph.D. gonzalosanchez411@gmail.com

1 Sanchez GM and Meltzer ES. The Edwin Smith Papyrus – Updated Translation of the Trauma Treatise and Modern Medical Commentaries. Lockwood Press. Atlanta Ga. 2012.

On 2014 Mar 20, Jonathan Eisen commented:

I wrote a brief post about this paper / software for my "Tree of Life" blog. See A little bit about PhyloSift: phylogenetic analysis of genomes and metagenomes.

On 2014 Feb 15, Amanda Capes-Davis commented:

None

On 2014 Feb 15, Amanda Capes-Davis commented:

Please be aware that INT-407 cells are not intestinal epithelial cells. The cell line is known to be cross-contaminated and is actually HeLa. A list of known cross-contaminated or other misidentified cell lines can be found at http://iclac.org/databases/cross-contaminations/.

On 2016 Jan 19, Diana Petitti commented:

Financial Disclosure I was asked to review this publication by the American Suntanning Association. I was compensated for my time in conducting this review and in preparing a report based on the review. The American Suntanning Association did not have rights to comment on these comments or to modify the final report.

Scope of Comment These comments summarize my conclusions with regard to the estimates of the prevalence of ever exposure to indoor tanning in adults. These prevalence estimates are key inputs in the model used to estimate the number of skin cancers attributable each year to indoor tanning in United States, Northern and Western Europe, and Australia.

Description of Systematic Review Eligibilty Wehner et al. (2104) state that their systematic review sought to obtain prevalence estimates "representative of the general population." They do not specify the criteria used to define an estimate of prevalence representative of the general population. The e-Appendix description of the studies deemed eligible does not provide detail on the sampling frame/study methods or response rates.

My Review I read the full text for all but one of the 17 publications that Wehner et al. (2014) identified as reporting estimates of the prevalence of ever exposure to indoor tanning in adults. The publication for which full text could not be retrieved (Mawn and Fleischer 1993; Wehner reference 23) provided detailed information on the study population in its abstract. I evaluated the accuracy/credibility of Wehner et al.’s meta-analytically derived estimates of the prevalence of ever exposure to indoor tanning in adults in the United States, Northern and Western Europe and Australia considering whether the studies were based on data representative of the general population. The accuracy/credibility of these estimates determines the accuracy/credibility of Wehner et al.’s model-based estimates of the number of skin cancers attributable each year to indoor tanning.

My Findings United States

None of the studies reporting the prevalence of ever exposure to indoor tanning in adults that Wehner et al. 2014 identified in their systematic review provide data representative of the general adult population of the United States. Several of the studies are from haphazard samples. For example, one study, Mawn and Fleischer 1993 (Wehner et al. reference 23), collected data using self-administered questionnaires distributed to “477 persons in a shopping mall, at a social gathering, and on a vacation cruise ship.” Another study, Hoerster et al. 2007 (Wehner reference 40), collected data about the prevalence of ever exposure to indoor tanning in adults in the United States from a telephone survey of households that were selected because they had a high likelihood of having a child 14, 15, 16, or 17. Responses about ever exposure to indoor tanning in adults pertain to households with an adult who had a child age 14, 15, 16, or 17 years. One study, Lazovich et al. 2008 (Wehner reference 36), collected data about the prevalence of ever exposure to indoor tanning in adults in the United States using an interviewer-administered questionnaire given to a 26 adults recruited from an undergraduate psychology seminar and a convenience sample of adult staff and friends in Virginia and from flyers, announcements, and advertisements in Massachusetts. One study Cohen et al. 2013 (Wehner reference 29) collected data about the prevalence of ever exposure to indoor tanning in adults in the United States using a self-administered questionnaire given to a “convenience” sample of 100 parents of children being seen in three pediatric practices in Chicago.

One study, Mawn and Fleischer 1993 (Wehner et al. reference 23), collected data in 1992, more than two decades before 2014, the year for which the estimate of the prevalence of ever exposure to indoor tanning in adults was made. Several other studies collected data more than a decade before 2014.

The meta-analytically derived estimate of the prevalence of ever exposure to indoor tanning for adults in the United States based on the studies identified by Wehner et al. (2014) is meaningless; the estimate of the number of skin cancers attributable to indoor tanning in the United State based on this meaningless estimate is meaningless.

Northern and Western Europe

The Wehner et al. (2014) systematic review identified studies of the prevalence of ever exposure to indoor tanning adults that were done in the United Kingdom, Ireland, France, Germany, Denmark, and Sweden. Only one study, Borner et al. 2009 (Wehner reference 27), had a sampling frame that could have yielded data representative of Germany but the r response rate was very low (13%). Germany is not representative of all of Northern and Western Europe. Austria, Belgium, Luxembourg, the Netherlands, Estonia, Finland, Iceland, Latvia, Lithuania, Norway and Switzerland are countries in Northern and Western Europe for which no prevalence data were identified.

One study, Bränstrom et al. 2004 (Wehner reference 28), collected data about the prevalence of ever exposure to indoor tanning in adults based on population-based sample limited to adults age 18-37 years in Stockholm County, Sweden. One study, Pertl et al. 2010 (Wehner reference 37), collected data about the prevalence of ever exposure to indoor tanning in adults using an interviewer-administered questionnaire given to “convenience sample” of adults between age 16 and 27 recruited in “various locations around Ireland (e.g., schools, sports clubs, universities and train stations.)”

One study, Jackson et al. 1999 (Wehner reference 33), collected data in 1995, nineteen years before 2014, the year for which the estimate of prevalence was made. Several other studies collected data more than a decade before 2014.

The meta-analytically derived estimate of the prevalence of ever exposure to indoor tanning for adults in Northern and Western Europe based on the studies identified by Wehner et al. (2014) is meaningless; the estimate of the number of skin cancers attributable to indoor tanning in Northern and Western Europe based on this meaningless estimate is meaningless.

Australia

The Wehner et al. (2014) systematic review identified one study (Francis et al. 2010; Wehner reference 31) that reported a measure of the prevalence of ever exposure to indoor tanning adults in Australia that is probably “in the ball park.” The prevalence measure based on data collected in 2007/2008 is reasonably current considering 2014 as the year for which the estimate was made. The sources of data on the annual number of incident melanoma and non-melanoma skin cancers in Australia is credible and I was able to verify the accuracy of these estimates.

On 2014 Mar 10, Madhusudana Girija Sanal commented:

In 1894 when Jacques Loeb occasionally observed the formation of a large blob in some of the early embryos when he attempted to induce parthenogenesis in sea urchin embryos using different salt concentrations. He found some unique properties with this blob. I think he observed pluripotency. However, mammalian cells are light years away from sea urchin cells. The authors of the STAP cells spent a lot of time on sophisticated tests like tetraploid complementation assay but skipped several basic experiments- what happens immediately, after 12h, 24h, 48h, 72h to the cells in terms of gene expression, cell physiology, signalling. This is a problem of the peer review system- reviewer has to believe what the 'researchers' present before them. Indeed, no reviewer can request an independent agency to repeat a critical part or suspecious part of the experiment prior to publication! A critical exploit! Even legends like Yamanaka, Rossant seems to agree than disagree with the results (interviews).

On 2014 Feb 24, Alexis Verger commented:

The raw sequencing data are now available :

http://www.ncbi.nlm.nih.gov/Traces/sra/?study=SRP038104

http://www.ncbi.nlm.nih.gov/sra/?term=SRP038104

On 2014 Mar 10, Madhusudana Girija Sanal commented:

It is a common observation that stressed cells, dying and dead cells start to become fluorescent. So early observation of florescence as a pointer to pluripotency (Oct4-GFP) needs to be substantiated by other experiments. The authors need to provide more evidence (molecular biological and morphological) and details (protocol) regarding STAP during the early hours to days post-stress, because this period seems to be the most critical in reprogramming (compared to characterization of the final product-STAPs). The authors underscore that the production of STAP is not the result of any type of “selection” during culture. However in a subsequent protocol paper (doi:10.1038/protex.2014.008) they see a ‘possibility of negative cell-type-dependent bias’. In the new protocol paper they also didn’t observe any T-Cell receptor rearrangement. So what type of PTPRC (CD45+) cells contributed to STAP? It is also difficult to understand or define the difference between STAP cells and STAP stem cells.

On 2014 Feb 06, ERIC GREENE commented:

See this educational video describing how Cas9 searches DNA for target sites: http://www.youtube.com/watch?v=M739wgbcKuA

On 2014 Apr 23, Janet Kelso commented:

We, the authors of the Neandertal genome papers, are not aware of any modern human contamination over and above that which is clearly quantified and reported in our manuscripts. We have developed and published a number of approaches to quantify modern human contamination using both mitochondrial and nuclear sequence data (Green RE, 2008, Green RE, 2009, Meyer M, 2012, Prüfer et al., 2013). In each paper where we report DNA sequences from archaic humans we have carefully quantified and reported any modern human contamination detected. As we have said in responsee to previous similar comments, it is difficult to respond to rumors without knowing the substance of the analyses on which such rumors are based. We would be interested to see the data and analyses that suggest modern human contamination in the in excess of that reported in the published Neandertal genome manuscripts.

On 2014 Feb 24, Steven Salzberg commented:

These authors identified human variants that matched the Neandertal reference genome, and used this as the basis (in part) for identifying extensive evidence of Neandertal remnants in modern humans. However, I have heard recently, from geneticists working in the area, that some of the published Neandertal sequences contain human contaminants, despite the efforts of the Paabo group to keep such contaminants out. If so, then building additional conclusions such as this paper on the "Neandertal" edifice might turn out to be a house of cards. I think more work is needed to ascertain that sequences published as Neandertal are truly free of modern human sequences.

On 2014 Feb 17, Reinhard Stindl commented:

Vernot et al. discovered over 15 Gb of introgressed Neandertal sequence in modern humans, but only 23 Mb of introgressed sequence per individual (on average). Very interesting results indeed! Surprisingly, no modern human sequences have been found in genome drafts of Neandertals despite the claim of multiple hybridizations. Green RE, 2010 Wills C, 2011 It remains to be seen, if other Neandertal sequencing projects will discover genetic remnants of hybridizations with modern humans. If not, the current model of human evolution is in trouble. The multiregional model with local Neandertals directly transforming into modern humans might better explain the extreme variation of inherited Neandertal sequences in modern human populations and the absence of any modern human genetic sequence in Neandertal DNA.

On 2014 Nov 26, Guillaume Filion commented:

This article is one of the "CISCOM meta-analyses", which are very similar papers written by different authors. For more information about the CISCOM meta-analyses, check the blog post "A flurry of copycast on PubMed" at the following link http://blog.thegrandlocus.com/2014/10/a-flurry-of-copycats-on-pubmed

On 2014 Mar 07, Jonathan Eisen commented:

One of the authors of this article - Chase Beisel - wrote a guest post on my blog discussing the "Story behind the paper" for this article. See his post here

On 2014 Feb 23, Ferenc Zsila commented:

Although the tocainide analogues studied here bear an asymmetric center (see in Fig. 1), their stereochemical characterization is missing. It remains unclear whether racemic or optically pure samples were used for the binding experiments.

Due to the presence of the chiral carbon atom, enantiomers of these molecules display CD activity below 300 nm which may overlap with the induced CD signals of the RSA-bound site markers used in the study (see the induced CD curves of diazepam, phenylbutazone, and ketoprofen in Fig. 6-9). Such a mutual spectral perturbation makes the reliable interpretation of the CD displacement results difficult. Unfortunately, CD spectra of free and albumin-bound forms of the enantiopure tocainide analogues are not presented.

The use of salicylate as the marker of Sudlow's site I (subdomain IIA) is ambiguous. It is a tiny molecule which can associate to multiple sites as demontrated by crystallographic results which revealed at least three, multidomain binding positions of diiodosalicylic acid on fatty acid-free bovine and equine serum albumin (Sekula B, 2013). Furthermore, the drug binding cavity of subdomain IIA is large enough to simultaneously accomodate two ligand molecules (Ghuman J, 2005). Thus, tocainide and its derivatives may co-bind with salicylate to this site. The same is hold for the case of phenylbutazone.

According to the Experimental Section, fatty acid-free RSA was used in all experiments. It is proposed that the primary binding site of tocainide analogues is located in subdomain IIIA which hosts one of the three highest-affinity binding sites of dietary fatty acids. Since fatty acids are the most abundant physiological ligands of serum albumin (Simard JR, 2006), fatty acid displacement measurements should also been performed to obtain more realistic data on the RSA binding behaviour of tocainide analogues.

In the course of mapping potential binding sites, the authors focuses exclusively on the classical Sudlow sites located in subdomain IIA and IIIA, respectively. Thus, they completely ignore recent findings showing the existence of a third primary drug binding area within subdomain IB (Zsila F, 2013). Biliverdin is the specific CD marker of this site which could be used for testing the subdomain IB binding of tocainide derivatives. This would be important since X-ray crystallographic studies verified the accomodation of the structurally related lidocaine molecule at the open entrance of the large binding crevice in subdomain IB (Hein KL, 2010).

Neither RSA nor HSA affinity constants of the tocainide analogues are reported in the paper. The number of the binding sites is also missing.

As it is claimed in the Conclusions, the tocainide analogues "showed ... a competitive behaviour with diazepam and bilirubin". Taking into consideration the very high-affinity RSA binding of bilirubin (Ka = 2-4 x 10⁶ M^-1 ) compared to the weak RSA association of the analytes (Ka ~ 10⁴ M^-1 ) this postulation is rather questionable. Additionally, in a recent work the primary RSA binding site of bilirubin has been assigned to subdomain IIA (Goncharova I, 2013), which was excluded by the present study as a possible binding locus of tocainides: "The addition of increasing concentrations of compounds BO3 and BO14 up to a molar ratio [competitor]/[marker] 8/1 determined a blue-shift of the induced CD spectrum of [RSA]/[PBU] 1/1 complex (Fig. 6). This result suggests an allosteric interaction between the site I marker and competitors, without a significant displacement of the marker, consistently to the results obtained by affinity chromatography." (p. 9).

On 2014 Feb 05, Anders von Heijne commented:

These findings are in line with our experience in using DTI in comatose patients after cardiac arrest. Eigenvalue maps are often very helpful as they show the extent of white matter injury most clearly. It is always helpful to have eigenvalue maps reconstructed when reading clinical DTI, in order to dechipher any FA-changes.

On 2014 Apr 25, Attila Csordas commented:

proteomics data available via PRIDE/ProteomeXchange http://www.ebi.ac.uk/pride/archive/projects/PXD000917

On 2017 Apr 10, Harri Hemila commented:

Problems in the review on the common cold

Allan and Arroll misrepresent the findings of the Cochrane review on vitamin C and colds by Hemilä H, 2013, see CMAJ eLetter.

Allan and Arroll refer to 2 reviews on zinc and the common cold Singh M, 2011 and Science M, 2012, but overlooked severe problems which had been identified with these reviews, see HDL and HDL and CMAJ eLetter.

On 2017 Jul 28, Randi Pechacek commented:

Jonathan Eisen mentioned this article on a microBEnet blog post while discussing the study of microbiomes of the built environment. Bubba Brooks, first author of this paper, also wrote a blog post on microBEnet that gives some background into his personal life and inspiration for this paper.