This retrospective study investigated whether a full-component chart-derived Lee score is associated with upfront selection of androgen receptor signaling inhibitors (ARSI) and with competing (other-cause) mortality among men with metastatic hormone-sensitive prostate cancer (mHSPC). The authors aimed to determine if a chart-based geriatric metric could inform treatment-intensification decisions in this population.
From an initial pool of 136 potentially eligible patients receiving docetaxel-free first-line therapy for mHSPC, 96 patients were included who had direct pretreatment documentation of all four functional items required to compute the full 12-component Lee score. The study used the original point allocations for the 12-component score and analyzed the score as a continuous variable. Forty patients were excluded due to incomplete functional documentation; the manuscript reports differences between included and excluded patients in age, albumin, ARSI use, and observed follow-up.
Upfront ARSI receipt was evaluated with adjusted Firth logistic regression to account for small-sample bias in logistic estimates. Competing-mortality outcomes were assessed using univariable cause-specific Cox proportional hazards models and Fine–Gray subdistribution hazard models. The reverse Kaplan–Meier method provided the median estimate of follow-up duration.
In the item-complete analysis cohort, median age was 80 years and the median chart-derived Lee score was 11. Of the 96 included patients, 33 (34.4%) received upfront ARSI. The reverse Kaplan–Meier estimate of median follow-up was 37 months. During follow-up there were 29 deaths, of which 17 were attributed to other causes (non–prostate cancer mortality).
In the adjusted Firth logistic regression model (reported for n = 91), a higher Lee score exhibited an inverse association with receipt of upfront ARSI, but the association was uncertain: odds ratio per point increase in score was 0.84 (95% confidence interval 0.69–1.01; p = 0.066). The point estimate suggests that higher frailty or comorbidity burden as captured by the Lee score may reduce the likelihood of treatment intensification with ARSI, but the confidence interval and p-value indicate the result did not meet conventional thresholds for statistical significance and should be interpreted cautiously.
In univariable cause-specific Cox analysis, the chart-derived Lee score was associated with other-cause death: hazard ratio per point was 1.16 (95% CI 1.02–1.32; p = 0.028). The Fine–Gray competing-risks model produced a similar estimate. These results imply that higher scores—reflecting greater comorbidity, functional impairment, or frailty—were linked to increased risk of non–prostate cancer mortality in this cohort.
The authors emphasize several important limitations. The primary analyses were conducted in an item-complete subset (96 of 136 potentially eligible patients), and excluded patients differed from included patients in several baseline variables (age, albumin, ARSI use, and follow-up). The number of competing events was relatively small (17 other-cause deaths), which limits precision and may increase the risk of chance findings. The borderline association between score and upfront ARSI receipt (p = 0.066) was described as uncertain rather than definitive. Given these constraints, the authors characterize the results as exploratory and hypothesis-generating rather than conclusive.
In this retrospective mHSPC cohort, a higher chart-derived Lee score was associated with greater risk of other-cause mortality and showed a possible inverse relationship with treatment intensification using upfront ARSI. The study underlines that direct functional assessment should be used to individualize treatment decisions in older patients with metastatic prostate cancer. Because the analysis was limited to an item-complete subset with relatively few competing events, the findings require confirmation in larger, prospectively assessed cohorts before they can directly guide treatment-selection algorithms.
Keywords reported by the authors included Firth logistic regression, androgen receptor signaling inhibitor, competing risks, geriatric assessment, and prostate cancer. The authors note the exploratory nature of the work and recommend direct functional assessment to inform individualized treatment choices in older men with mHSPC.