Abstract 655P
Background
Clinical trials have reported conflicting findings on the role of D plus ADT in p with mCSPC according to extent of disease. We have retrospectively analyzed the impact of extent of disease on progression-free survival (PFS) and overall survival (OS) in p with mCSPC treated with D plus ADT or ADT alone in clinical practice.
Methods
Between 2015 and 2019, 160 p with mCSPC were treated at centers of the Catalan Institute of Oncology (Catalunya, Spain). For the present study, we have classified these p according to extent of disease. Those with ≥five bone metastases and/or visceral metastases were defined as “high-volume” (HV; n=87), and all other p were defined as “low-volume” (LV; n=73).
Results
One hundred p (60 HV; 40 LV) received D plus ADT and 60 p (27 HV; 33 LV) received ADT alone. Median age was 69.5 years; 79% of p were ECOG PS 0-1; 81% had Gleason 8-10. Bone, lymph node, and visceral metastases were present in 87%, 66%, and 15% of p, respectively. p receiving ADT alone were older with poorer ECOG PS than p receiving D plus ADT. 95 p (59%) progressed to castration-resistant prostate cancer, 74% of whom were then treated with abiraterone or enzalutamide and 24% with D or cabazitaxel. Median PFS was 18.9 months (m) in the 100 p treated with D plus ADT and 13.4 m in the 60 p treated with ADT alone (P=0.01). Among LV p, PFS was longer in those receiving D plus ADT than those receiving ADT alone (29.6 m vs 15.0 m; P=0.002). In contrast, in HV p, no differences in PFS according to treatment were observed. Although OS was slightly longer in p treated with D plus ADT (39.2 m vs 35.7 m, p=0.26), the difference was not significant either in HV p (P=0.41) or LV p (P=0.11). Multivariate analyses identified advanced age, PS 2, HV, and ADT alone as markers of shorter PFS and only PS 2 as a marker of shorter OS.
Conclusions
Despite the limitations inherent in a retrospective, non-randomized study, our findings suggest a benefit in PFS for D plus ADT in LV p but not in HV p. A longer follow-up and marginal structural modeling are warranted to determine the impact on OS in both HV and LV p.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.