Shreyas Kalantri, Shiva Balasubramanian, Jincong Q Freeman, Tyler Jones, Priyanka Sheth, Srinidhi Karthik, Jason Chesney, Rohit Kumar
QoL end points in GU cancer trials are frequently planned but selectively reported, with findings consistent with result-dependent disclosure. This pattern may compromise value assessment frameworks and shared decision making. Enforcement of CONSORT-PRO, SPIRIT-PRO, and SISAQOL standards is needed.
PURPOSE: Quality-of-life (QoL) end points are essential for treatment decisions and value assessment in genitourinary (GU) cancer, yet the extent of QoL planning-reporting concordance and outcome reporting bias is unknown. We evaluated the frequency, correlates, and concordance of QoL end point planning and reporting in phase III GU cancer trials.
METHODS: We conducted a systematic review of phase III randomized clinical trials evaluating systemic therapies for GU malignancies published from 2020 through 2024 (PubMed, EMBASE). QoL prespecification was ascertained from ClinicalTrials.gov or protocols. Associations with QoL reporting were assessed using univariable and multivariable logistic regression.
RESULTS: Among 80 clinical trials, QoL was prespecified in 68 (85%) yet reported in only 38 (47.5%). Of the 68 trials that prespecified QoL, 30 (44.1%) did not report results. Positive primary end point results were the strongest correlate of QoL reporting (63% v 25%; adjusted odds ratio [OR], 4.84 [95% CI, 1.74 to 14.60]; P = .003). Single-country scope was independently associated with lower QoL reporting (OR, 0.22 [95% CI, 0.04 to 0.83]; P = .04). Among trials that prespecified QoL, the association with result direction persisted (71% v 31%; P = .001), and neither funding source nor journal impact factor explained the discrepancy. QoL reporting rates were similar from 2020 through 2023 (50%-61%) and dropped to 11% in 2024; in a sensitivity analysis modeling publication year categorically, only 2024 differed significantly from 2020 (OR, 0.07 [95% CI, 0.01 to 0.34]; P = .003). No trial designated QoL as a primary end point.
CONCLUSION: QoL end points in GU cancer trials are frequently planned but selectively reported, with findings consistent with result-dependent disclosure. This pattern may compromise value assessment frameworks and shared decision making. Enforcement of CONSORT-PRO, SPIRIT-PRO, and SISAQOL standards is needed.