#TeachSpeak: reverse causation

Are you a knee-jerker or a deep-thinker? #TeachSpeak #MSBlog

Yesterday's post on B-cells and breast cancer has generated a lot of debate, some justified and some not. Before joining the debate you need to know how the mind works. There is a rapid, or fast, intuitive decision making system called system 1 and a slow more rational decision making system called system 2. If you are interested in reading about the way we think I would suggest you read 'Thinking, Fast and Slow', the best-selling book by Nobel laureate Daniel Kahneman. Just reading the introductory chapter of the book will be enough to give you the gist of system 1 and system 2 thinking. 

Based on Kahnerman's theory a system 1 (knee-jerk) response to my post is that B-cells prevent breast cancer hence ocrelizumab causes breast cancer. A system 2 response, the slower more rational response, would lead to a more nuanced interpretation. Firstly, you need to be aware that greater than 50% of oncology research findings can't be reproduced. In other words this paper's finding are likely to be a false positive based on pure statistics. 


Then there is 'reverse causation', which is likely to be the best explanation. In other words benign breast lesions that don't recruit B-cells are different biologically to those that do and are more likely to become malignant. The B-cells may simply be innocent bystanders and their presence or absence makes no difference in the long-term. Despite this there is the possibility that the findings are causal, i.e. B-cells are involved in preventing breast cancer. 

So what needs to be done? Other groups need to try and replicate the Mayo clinic study's findings and additional studies need to be done to explore the role of B-cells in breast cancer pathogenesis. We also need to set-up and monitor breast cancer incidence in relation to B-cell depletion therapies to assess whether or not the signal in the ocrelizumab trial programme is real or not. As with most DMTs the trial programme is too small and too short to give a definitive answer. All I would suggest is that pwMS considering ocrelizumab are made aware of the uncertainty and let them make up their own mind about the potential risk. You also have to remember that breast cancer can be detected early with increased vigilance and that the prognosis of breast cancer in 2017 is very good most patients with breast cancer are cured. All pwMS, women and men, need to be made aware about the possibility of secondary malignancies on immunosuppressive  therapies and taught how to monitor for possible malignancies. This message is not unique to ocrelizumab and applies to all most all of the DMTs. 

Another example of reverse causation in MS is low vitamin D levels in pwMS that have been linked to MS disease activity. Because of this observation everyone has interpreted the results that vD supplementation is essential to control MS disease activity. In reality almost every inflammatory disease studied has shown low vD levels that are linked to the degree of inflammation. The most likely explanation is a consumptive hypovitaminosis  D; i.e. inflammatory cells consume vD as part of their biology. Giving vD is unlikely to make any difference to their functioning or the outcome of the diseases concerned. We did a meta-analysis of the trials of vD supplementation that had been done to date and found no indication of vD being a DMT. Please note this comment does not apply to vD supplementation and MS prevention. Based on the literature and epidemiological studies vD plays a role in MS risk and there is an overwhelming case for doing vD prevention studies. 

Baker & Dolgin. Cancer reproducibility project releases first results. Nature 18 January 2017.

Excerpts

..... The Reproducibility Project: Cancer Biology launched in 2013 as an ambitious effort to scrutinize key findings in 50 cancer papers published in Nature, Science, Cell and other high-impact journals. It aims to determine what fraction of influential cancer biology studies are probably sound — a pressing question for the field. In 2012, researchers at the biotechnology firm Amgen in Thousand Oaks, California, announced that they had failed to replicate 47 of 53 landmark cancer papers. That was widely reported, but Amgen has not identified the studies involved.

..... The reproducibility project, by contrast, makes all its findings open — hence Ruoslahti’s discomfort. Two years in, the project downsized to 29 papers, citing budget constraints among other factors: the Laura and John Arnold Foundation in Houston, Texas, which funds the ­project, has committed close to US$2 million for it. Full results should appear by the end of the year. But seven of the replication studies are now complete, and eLife is publishing five fully analysed efforts on 19 January.

...... These five paint a muddy picture (see ‘Muddy waters’). Although the attempt to replicate Ruoslahti’s results failed, two of the other attempts “substantially reproduced” research findings — although not all experiments met thresholds of statistical significance, says Sean Morrison, a senior editor at eLife. The remaining two yielded “uninterpretable results”, he says: because of problems with these efforts, no clear comparison can be made with the original work.

CoI: multiple

Labels: , , , , , ,