Talk:TREM2

Wiki Education Foundation-supported course assignment
This article was the subject of a Wiki Education Foundation-supported course assignment, between 7 September 2021 and 13 December 2021. Further details are available on the course page. Student editor(s): Ryanpm11.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 10:38, 17 January 2022 (UTC)

Clinical significance
Hi. The secondary sources that I have provided document a strong associate between rare variants of the TREM2 gene and AD in humans, not animals. In addition, this is not a clinical trial of a drug, but rather an "experiment of nature" where a rare mutation naturally occurs in humans. Obviously it would be completely unethical to perform a double blinded trial where a mutation in human embryos is performed on half the subjects and wait 80 years to see if they develop AD. Finally, the adjective "strong" is not POV. It is the adjective that is used in the secondary source to describe the association. Genetic factors, such as rare variants in the microglial-expressed gene TREM2, strongly impact the lifetime risk of developing AD. If I am guilty of anything, it is close paraphrasing. Boghog (talk) 16:14, 19 May 2021 (UTC)

Absence of human clinical effect
These two edits were justified because the only evidence of TREM2 having a role in the pathogenesis of Alzheimer's disease is from in vitro, lab animal studies, or post-mortem sample analysis - evidence that is not "clinical significance". This is speculative and unencyclopedic per WP:NOTJOURNAL #6-8. At best, the evidence for TREM2 being involved is coincident, and no cause-and-effect relationship has been proven, leaving the evidence as only conjecture on some future understanding, WP:CRYSTAL.

From the 3 references provided,  the authors use such speculative terms as "possible relationship", "suggest", TREM2 mechanism of action is undefined, an "hypothesis", etc. Although extensive and promising, the research on TREM2 remains at a hypothetical lab level which is not at a standard of "clinical significance", failing WP:MEDASSESS. Zefr (talk) 16:41, 19 May 2021 (UTC)


 * While the exact mechanism may not be completely worked out, the genetic association is strong. Therefore the phrase clinical significance is justified. Full stop. Boghog (talk) 16:47, 19 May 2021 (UTC)
 * in vitro – no, in vivo; lab animal studies – no, human; post-mortem sample – this is the gold standard method of an AD diagnosis. Boghog (talk) 16:53, 19 May 2021 (UTC)
 * Unacceptable, low-quality evidence within the category of primary research, left pyramid of WP:MEDASSESS. Too preliminary and too non-neutral to be "clinical" or "significant". I'm ok with the sentence and sources as research, but the subhead "Clinical significance" is misleading to the general user and to clinical science and practice. Zefr (talk) 18:20, 19 May 2021 (UTC)
 * There is a crystal clear association between TRIM2 mutations and clinical effect (AD) supported by multiple secondary sources. low-quality evidence within the category of primary research All evidence is initially primary.  If there is a consensus conclusion from secondary sources, as there is in this case, then it passes WP:MEDRS.  Therefore the categorization of the TREM2 disease linkage as "clinical significance" is entirely appropriate. Too preliminary? Too non-neutral? How so? You do understand this is not a treatment. Right? At best, it might be used a prognostic, but that has not been claimed in the text. These sources also talk about the mechanism of how TRIM2 affects AD and a lot of that research is more speculative.  But again, the current "clinical significance" section does not mention anything about mechanism. So what is the problem? Boghog (talk) 18:50, 19 May 2021 (UTC)
 * WP:NPOV states ... representing fairly, proportionately, and, as far as possible, without editorial bias, all the significant views that have been published by reliable sources .... The reliable source has used the adverb "strongly", so including in this article in no way violates the NPOV policy.  In fact, one could argue by not including the adverb "strongly" waters down the conclusion and therefore is biased in the opposite direction.  Also you need to read the sources more carefully.  Some of the conclusions, for example on mechanism, are only based on in vitro and animal data and these conclusion are appropriate for a "research section".  However the conclusion on the linkage TREM2 mutations and AD is based on human data and because this conclusion is made by reliable secondary sources, it is appropriate to put this under a "clinical significance" heading. Boghog (talk) 03:25, 21 May 2021 (UTC)
 * Bottom line for convincing about "clinical significance" or certainty of cause-and-effect pathological mechanism would be that this association of TREM2 mutations with diagnosed diseases would be in a systematic review in a reputable clinical journal, or stated by a clinical neurological organization in a clinical practice position statement (WP:MEDSCI and WP:MEDASSESS). I find no evidence for those conditions. With the one exception of the J Clin Invest reference (current #9), all the sources in this article are preclinical research publications. Note also that perhaps the most thorough review is in a Frontiers-MDPI journal which is suspect for editorial quality and possible predatory publishing. Zefr (talk) 13:49, 21 May 2021 (UTC)
 * . The associations between TRIM2 mutations and AD are based solely on observations of human populations.  They were not based on in vitro or animal studies, only on human observational studies.  Hence it it justified to put these associations under the "clinical significance section". Boghog (talk) 05:45, 21 May 2021 (UTC)
 * Your edits added supporting references, but the subhead "clinical significance" overstates the detail of the sentences, which describe "associations" with disease risk or onset. On Clinicaltrials.gov, I found 3 early-stage 2020-21 trials - all still recruiting - where TREM2 is among several biomarkers being evaluated for AD diagnosis, i.e., it is not a mainstay marker, but is among potentially useful markers, and therefore would not be considered "significant" yet, as there is no trial evidence or MEDSCI review for that. With this edit, I changed the subhead for consistency, and am satisfied we can end this lengthy discussion mainly over semantics. Zefr (talk) 13:49, 21 May 2021 (UTC)
 * You keep bringing up side issues like biomarkers, mechanism, etc. and then conclude statements about disease association in the "clinical significance" section are somehow weakened when these side issues are not even mentioned in the section. That is an association fallacy. A validated biomarker or well established mechanism would strengthen the argument, but are not necessary to make the argument.  Also phrase "association" is a bit fuzzy and covers the range from borderline statistically significant correlations to established causal relationships.  The TREM2 R47H variant shows highly significant association with Alzheimer's disease (P<0.001) thus justifying the adverb "strongly" associated.  All patients with homozygous loss of function TREM2 mutants will develop Nasu-Hakola disease in their 40s.  Hence this association is causal and can justifiably be called "clinically significant".  Systemic reviews or position statements are unlikely to be ever written about this orphan disease, and I therefore I think you are applying an unreasonable standard to require such sources. A review article in a respected journal is sufficient to make a medical claim, unless of course, it is an extraordinary claim or a claim that effects a large number of people, which this claim is not and does not. Boghog (talk) 14:58, 21 May 2021 (UTC)
 * You are arguing a point that no clinical organization or WP:MEDRS review states, making this an easy position for editors to write for the article's general users - WP:NOTJOURNAL #8. TREM2 is not yet recognized by medical experts as "clinically significant" or there would be reviews in medical journals about it. Last comment for me. Zefr (talk) 15:09, 21 May 2021 (UTC)
 * And no clinical organization ever will, because it is an orphan disease that affects a very small number of people (something like 200 worldwide). Again, your standards are unreasonable. Both associations are clinically significant because they are based on observations of human populations and have been review in reliable secondary sources. Boghog (talk) 15:25, 21 May 2021 (UTC)