blog

(This is part 2 of a series of posts featuring speakers from “Challenging the Science Publishing Status Quo”, an evening of talks about peer review, data sharing, and open access. Previously: Lawrence Kane on rapid publication.)

Keith Flaherty is Director of Developmental Therapeutics at the Massachusetts Cancer Center. In his talk he addressed the benefits of publishing negative results. He focused his talk on therapies for cancer, in particular melanoma, since that’s the area he is most familiar with.

Flaherty’s slides are available via F1000Posters. Click the links in the text below to see the video segments.

In the first part of his talk, he described how in many melanoma patients who were using BRAF-targeted therapies, the treatment lost effect after a while. A study into the mechanism of resistance to BRAF-targeted therapies suggested that it was caused by upregulation of the insulin-like growth factor receptor. This was validated in human tumours – but only from five patients, and only one of them showed the expected effect. Meanwhile, several other groups were unable to reproduce the initial experiment, and none of those studies were published. (The initial IGFR upregulation was thought to be caused by growing the cell line in insulin-containing growth media).

Flaherty pointed out that in fields like cancer, where some patients have no hope beyond a certain type of therapy, such as BRAF-targeted treatment, there is a drive to move to clinical studies as quickly as possible. Not having the negative data published right away means going forward with unrealistic expectations.

Next, Flaherty moved to preclinical studies, where negative data are still not being published. As a result, others will repeat the same studies, not knowing that it didn’t work the first time.

“The consequence [of not publishing negative results] is that multiple dead ends can be pursued by others who might develop an agent in the same class or aren’t aware of that effect, and all of the preclinical evaluation might ensue in the absense of that knowledge.”

And the availability of negative data doesn’t get better as you move further into patient studies. Flaherty mentioned that “nearly 10% of Phase III trials never made it into the public literature.”

In closing, he emphasized that publishing negative data in preclinical and clinical studies would save time and money, and could open up new pathways of investigation.

In the question round, after his talk, Flaherty addresses another important point: as a trainee, if you worked really hard but you ended up with negative data, not publishing that work makes it look like you didn’t do anything. Flaherty evaluates potential trainees by looking at their productivity, and being able to publish all your results – positive and negative – shows the full spectrum of work that was done.