Why do prostate cancer clinical trials (CT) discontinue prematurely in the era of COVID-19? An analysis of 559 CT from ClinicalTrial.gov

ABSTRACT

Background: The COVID-19 pandemic (C19P) is producing several detrimental effects on cancer care globally. CT play a decisive role to provide high quality literature evidence and “poor accrual” is the most common reason for their early discontinuation (ED). At our best knowledge, no data are available on ED of prostate cancer CT after the beginning of C19P. Material and methods: ClinicalTrial.gov was queried for terminated (T), withdrawn (W) and suspended (S) CT for the following terms “cancer”, “neoplasm” and “tumor”. CT not related on prostate cancer were excluded. The search was made for all the CT available from the inception to 26th February 2021, without any restrictions. The following characteristics were extracted reason for ED, study type (interventional [In] vs observational), sponsor (yes vs not). ED rate was compared between CT discontinued for C19P or not (χ2);p < 0.05 was set as statistically significant. A multiple linear regression analysis was also conducted to identify independent factors of ED. Results: 9990 CT were identified and 7901 CT were excluded because not related to prostate cancer. Thus, 559 CT were included 67% was T, 27% was W and 5% was S. Among CT classified as T, W and S, the frequency of In CT were 90%, 82% and 81% respectively, while the frequency of sponsored CT was 48%, 27% and 19% respectively. The most common reasons for ED were “poor accrual” (31%), “lack of funding” (7%) and “sponsor decision” (5%). No reason for ED was available for 13% of CT. Ten (2%) CT were discontinued for C19P (20% was T, 10% was W and 70% was S). Comparing CT discontinued due to C19P with those discontinued due to other causes, a lower rate of In-CT (88% vs 91%, p<0.05) was found in the C19P group. At the multiple linear regression analysis, it was found that C-19 was strongly positively correlated with ED (coefficient 0,62677, p<0.0001) whereas sponsored CT resulted negatively correlated with ED (coefficient -0,03717, p=0.0369). Conclusions: “Poor accrual” continues to be the main reason for ED of cancer CT, whereas C19P represents a new additional cause of ED. Sponsored trials showed less risk for ED. Further research is needed to maximize the expected benefit of cancer CT, reducing the anticipated risks.