Would you consider addition of docetaxel to abiraterone and ADT in metastatic hormone senstive prostate cancer patients who progress on ADT and abiraterone?

The question is incorrectly phrased, when a patient progresses on ADT and Abi they are not hormone sensitive.

In patients with high risk metastatic prostate cancer, I would not add docetaxel to abi and ADT yet as OS data has not matured. However, I do sequence them in a select group of high risk patients i.e., starting abi right after stopping 6 cycles of docetaxel.

The data from the PEACE-1 trial may end up helping to set a new standard for mHSPC and demonstrating safety of combination docetaxel with abiraterone, but if the question is whether or not to continue abiraterone in the setting of progression when starting docetaxel for mCRPC, PEACE-1 probably isn't going to answer that question.

However, a randomized phase 2 trial of 94 patients with CRPC by the Spanish Oncology Genitourinary Group (SOGUG) did look at a similar question. This was presented at GU ASCO 2020 in poster form. In this study, patients with CRPC were enrolled and treated with abi/pred, at progression were randomized to (A) continuation of abi/pred and adding docetaxel or (B) docetaxel alone. The primary endpoint was rPFS. There was no difference in rPFS or OS. Median OS was 17.3m and 16.9m respectively. The cohort which continued abi/pred had more hematologic toxicity. The study has not been published (to my knowledge), but it provides some data to suggest that there's no benefit of continuing abi at progression when planning to start docetaxel. 

https://meetinglibrary.asco.org/record/183558/abstract 

All this being said, I'm updating the above response based on GU ASCO 2022 results, as they seem relevant to this question.

At GU ASCO 2022, the results of the PRESIDE study were presented. PRESIDE was a randomized, double-blind, placebo-controlled, phase 3 study of continuing enzalutamide (or not) in chemotherapy-naïve, metastatic CRPC when switching to docetaxel. All patients enrolled (687) received ADT+enza and then the 544 patients had progression were then given docetaxel, and were randomized 1:1 to either continue enza during docetaxel, or to stop enzalutamide. PFS was significantly improved with continuation with enzalutamide (HR 0.72; 95% confidence interval [CI] 0.53, 0.96; p = 0.027), with a higher median PFS with enzalutamide (9.53 months) than with placebo (8.28 months). Treatment was well tolerated and AEs were consistent with its known safety profile. Though if you look at the KM PFS curves, there was not a huge difference in the curves. 

So back to the question above, there is now some data to support continuing NHT when patients progress and are planning to switch to chemotherapy.