Of mice and men: lost in translation

Abstract

The best-laid plans of mice and men often go awry.‘To a Mouse’ by Robert Burns 1786 Is it right to relegate clinical trials that don’t demonstrate clinical efficacy to lesser journals even though they are based on pre-clinical data published in high-impact journals? No. Trials specifically designed based on particular basic science work should be published as a ‘clinical trial update’ with the original pre-clinical work. Oncology as a field has a publication problem. Less than a fifth of clinical trials undertaken are actually published according to ‘Practicing on the Tip of an Information Iceberg’ [1.Ramsey S. Scoggins J. Commentary: practicing on the tip of an information iceberg? Evidence of underpublication of registered clinical trials in oncology.Oncologist. 2008; 13: 925-929Crossref PubMed Scopus (65) Google Scholar]. Many possible explanations exist for this publication bias with negative results as the primary driver. A comprehensive review of the proportion of oncology trials published and reasons for lack thereof revealed that 29% of completed clinical trials were not published due to lack of significant interest by the investigator, for either containing negative results or lacking relevancy to the field [2.Chapman P.B. Liu N.J. Zhou Q. et al.Time to publication of oncology trials and why some trials are never published.PLoS One. 2017; 12: e0184025Crossref PubMed Scopus (16) Google Scholar]. Most scientists and clinicians share the following belief: ‘It is likely that many unpublished studies contain important information that could influence future research and present practice policy’ [1.Ramsey S. Scoggins J. Commentary: practicing on the tip of an information iceberg? Evidence of underpublication of registered clinical trials in oncology.Oncologist. 2008; 13: 925-929Crossref PubMed Scopus (65) Google Scholar]. Many clinical trials are based on ‘promising’ pre-clinical work, especially in mouse models. These models, while convenient, often do not translate into clinical success. This is most evident in the fact that about half of all phase III trials do not reach their primary end point, with oncology being the largest contributor to this statistic [3.Arrowsmith J. Trial watch: phase III and submission failures: 2007–2010.Nat Rev Drug Discov. 2011; 10: 87.Crossref PubMed Scopus (347) Google Scholar]. Another complication is that even the pre-clinical studies themselves have a reproducibility problem [4.Begley C.G. Ellis L.M. Drug development: raise standards for preclinical cancer research.Nature. 2012; 483: 531-533Crossref PubMed Scopus (1881) Google Scholar]. They are often based on a single animal model and do not represent a target patient cohort. Herein, we propose that when clinical trial results become available, regardless if they are positive or negative, they should be published in the original journal that the pre-clinical work was published. Often, even if the pre-clinical work is published in a high-impact journal the clinical work may go unpublished or relegated to a poorly circulated journal, especially if the trial is negative. This frequently means that the corresponding clinical trial is not widely seen. This disturbs the checks and balances of our modern peer-review system [5.Prinz F. Schlange T. Asadullah K. Believe it or not: how much can we rely on published data on potential drug targets?.Nat Rev Drug Discov. 2011; 10: 712.Crossref PubMed Scopus (1245) Google Scholar]. It is our view that as an academic community we are obligated to provide a cohesive narrative of every scientific discovery, irrespective of species: mice or man. Much cancer research is funded by public taxes and philanthropic donations with the noble goal to end cancer. As a mark of honor and reverence to the patients who volunteered themselves for the trial, for the benefit of future patients, and as a part of the continuum in the scientific community, we propose that clinical studies should be published alongside their pre-clinical work, preferably as an ‘update’ due to the non-contemporaneous nature of pre-clinical data and the clinical trial. Herein, we narrate a case study that stirred us in part to write this. A pre-clinical study investigated the clinical activity of KIT inhibition with imatinib and immunotherapy on regression of gastrointestinal stromal tumors (GIST) [6.Balachandran V.P. Cavnar M.J. Zeng S. et al.Imatinib potentiates antitumor T cell responses in gastrointestinal stromal tumor through the inhibition of Ido.Nat Med. 2011; 17: 1094-1100Crossref PubMed Scopus (420) Google Scholar]. They used a mouse model of GIST to show that immune activation is a major mechanism of clinical response beyond KIT-driven cell proliferation. The major finding was that the combination of immunotherapy and imatinib may be better than imatinib alone. Our patient, a very motivated individual with KIT exon 11 mutant gastrointestinal stromal tumor, was refractory to tyrosine kinase inhibitors. He presented to our clinic, buoyed by immunotherapy news in other tumors, citing the pre-clinical paper. After enrollment on an ipilimumab and imatinib trial, he had rapid radiographic and symptomatic disease progression. The patient, spouse and their children were very distraught and raised the question of how valid a mouse model is in relation to human disease. Herein, the pre-clinical data inspired two phase I trials from major cancer centers to test the hypothesis that KIT inhibition and immune activation can have a synergistic effect in GIST. The studies enrolled 63 patients in total and were overall negative [7.Reilley M.J. Bailey A. Subbiah V. et al.Phase I clinical trial of combination imatinib and ipilimumab in patients with advanced malignancies.J Immunother Cancer. 2017; 5: 35.Crossref PubMed Scopus (46) Google Scholar, 8.D'Angelo S.P. Shoushtari A.N. Keohan M.L. et al.Combined KIT and CTLA-4 blockade in patients with refractory GIST and other advanced sarcomas: a phase Ib study of dasatinib plus ipilimumab.Clin Cancer Res. 2017; 23: 2972-2980Crossref PubMed Scopus (83) Google Scholar]. The biology of mouse and man is indeed a paradox. In this editorial, we presented a basic science paper, two clinical trials, and a case study. Since these trials were driven by academic investigators they were published and the investigators need to be commended. Regrettably, hundreds of negative studies go unpublished and the majority of them are industry driven [1.Ramsey S. Scoggins J. Commentary: practicing on the tip of an information iceberg? Evidence of underpublication of registered clinical trials in oncology.Oncologist. 2008; 13: 925-929Crossref PubMed Scopus (65) Google Scholar]. We outline some of the reasons negative trials are not published in Table 1 based on the literature and anecdotal experiences.Table 1Reasons for non-publication of clinical trials1.The clinical trial result was negative, or paradoxically opposite from pre-clinical data findings. Positive publication bias [2.Chapman P.B. Liu N.J. Zhou Q. et al.Time to publication of oncology trials and why some trials are never published.PLoS One. 2017; 12: e0184025Crossref PubMed Scopus (16) Google Scholar, 9.Callaham M.L. Wears R.L. Weber E.J. et al.Positive-outcome bias and other limitations in the outcome of research abstracts submitted to a scientific meeting.JAMA. 1998; 280: 254-257Crossref PubMed Scopus (223) Google Scholar, 10.Johnson R.T. Dickersin K. Publication bias against negative results from clinical trials: three of the seven deadly sins.Nat Rev Neurol. 2007; 3: 590-591Crossref Scopus (64) Google Scholar]2.Low clinical relevance/ unimportant topic as science moved forward [2.Chapman P.B. Liu N.J. Zhou Q. et al.Time to publication of oncology trials and why some trials are never published.PLoS One. 2017; 12: e0184025Crossref PubMed Scopus (16) Google Scholar]3.Investigator decision to pursue new projects over publishing negative study4.Decreased enthusiasm from the sponsor (decreased funding, decreased statistical support, economic impact of results etc.) [2.Chapman P.B. Liu N.J. Zhou Q. et al.Time to publication of oncology trials and why some trials are never published.PLoS One. 2017; 12: e0184025Crossref PubMed Scopus (16) Google Scholar, 11.Krzyzanowska M.K. Pintilie M. Tannock I.F. Factors associated with failure to publish large randomized trials presented at an oncology meeting.JAMA. 2003; 290: 495-501Crossref PubMed Scopus (271) Google Scholar]5.Decreased enthusiasm from investigator [2.Chapman P.B. Liu N.J. Zhou Q. et al.Time to publication of oncology trials and why some trials are never published.PLoS One. 2017; 12: e0184025Crossref PubMed Scopus (16) Google Scholar]6.Failure to write paper (lack of time, resources) [11.Krzyzanowska M.K. Pintilie M. Tannock I.F. Factors associated with failure to publish large randomized trials presented at an oncology meeting.JAMA. 2003; 290: 495-501Crossref PubMed Scopus (271) Google Scholar]7.Biased protocol with inappropriate biomarkers chosen for clinical convenience not reflecting the data from pre-clinical studies8.Small sample size [11.Krzyzanowska M.K. Pintilie M. Tannock I.F. Factors associated with failure to publish large randomized trials presented at an oncology meeting.JAMA. 2003; 290: 495-501Crossref PubMed Scopus (271) Google Scholar, 12.Massey P.R. Wang R. Prasad V. et al.Assessing the eventual publication of clinical trial abstracts submitted to a large Annual Oncology Meeting.Oncologist. 2016; 21: 261-268Crossref PubMed Scopus (26) Google Scholar]9.Low reader interest10.Publication could negatively affect investigator’s career in academia11.Never reached end point of study [3.Arrowsmith J. Trial watch: phase III and submission failures: 2007–2010.Nat Rev Drug Discov. 2011; 10: 87.Crossref PubMed Scopus (347) Google Scholar]12.Investigator left institution [2.Chapman P.B. Liu N.J. Zhou Q. et al.Time to publication of oncology trials and why some trials are never published.PLoS One. 2017; 12: e0184025Crossref PubMed Scopus (16) Google Scholar] Open table in a new tab

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 8 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.