How to choose duration of additional androgen deprivation therapy with salvage radiation therapy: short, long, more, or none?

Prostate cancer is the most common non-cutaneous cancers occurring in American men, and while most men with early-stage prostate cancers are cured, up to a third might manifest with biochemical recurrence (BCR) of prostate cancer. BCR is a disease entity which is characterized by a rising prostate-specific antigen (PSA) in the setting of a previously treated localized prostate cancer with either surgery or radiation therapy with curative intent. BCR is one of the disease states where there is a clear indication to obtain prostate-specific membrane antigen (PSMA) positron emission tomography (PET) scans to distinguish among localized, regional, and distant disease recurrence. For patients who are referred for curative intent salvage radiotherapy (sRT), a key question raised includes the role of additional hormone therapy. The choices include added androgen deprivation therapy (ADT) alone with a duration of 6 months (short-course) or 2 years (long-term) versus none or with additional androgen receptor pathway inhibitor (ARPI) intensification. While improvement in primary endpoints such as PSA recurrence with the use of short-course 6 months of ADT is seen, improvement in metastasis-free survival (MFS) or overall survival (OS) are seen with long-term 2 years’ duration of ADT with sRT, hence driving utility of added hormonal therapy. However, side effects and toxicity are equally important to consider when making treatment recommendations to patients.

Prostate cancer makes up the most common incidence of cancers in the USA in 2024,1 and about a third of men will present with BCR. The BCR definition varies depending on primary treatment of radical prostatectomy or primary radiotherapy. The American Urological Association (AUA) definition of BCR involves a consecutive PSA rise of ≥0.2 ng ml⁻¹ with a second confirmatory level of >0.2 ng ml⁻¹.2 There are also multiple diagnostic tools to help determine the presence of true local recurrence, distant metastatic disease, or occult metastatic disease, mainly in the form of imaging with PSMA PET scans. However, treatment with sRT offers the only potential chance of cure. The addition of systemic ADT potentiates the effects and improves outcomes. However, the exact duration of ADT in addition to sRT is a relevant clinical question (Table 1).

Table 1.

Select trials using androgen deprivation therapy with salvage radiation therapy

Trial name	Phase of trial/PSA level of patient population	Experimental arm	Control arm	Primary endpoint	Outcome	HR (95% CI; P)	Comment
RTOG 96013,4	Phase III/PSA 0.2 ng ml−1 to <4 ng ml−1	2-year bicalutamide + sRT	2-year placebo + sRT	OS	Improved OS with high PSA >1.5 ng ml−1	0.77 (0.59–0.99; 0.04)	>3-fold increase in high-grade cardiac and neurologic events and 2-fold increase in other cause mortality with 2 years of bicalutamide
GETUG AFU 169,10	Phase III/PSA 0.1 ng ml−1 to <2 ng ml−1	6-month goserelin + sRT	sRT	bRFS/bPFS	5-year bRFS/bPFS: 80% (95% CI: 75%–84%) of 6 months vs 62% (95% CI: 57%–67%) of no ADT	0.50 (0.38–0.66; <0.0001)	120-month PFS: 64% of 6-month ADT + RT vs 49% of RT (HR: 0.54, P<0.0001)
RADICALS-HD6	Phase III/PSA <5 ng ml−1	6-month ADT + sRT	sRT	MFS	10-year MFS: 79.2% of no ADT arm vs 80.4% of 6-month ADT	0.886 (0.688–1.140; 0.35)	No significant difference in toxicity > Grade 3: 17% of no-ADT vs 14% of 6-month ADT (P=0.15)
RADICALS-HD7	Phase III/PSA <5 ng ml−1	2-year ADT + sRT	6-month ADT + sRT	MFS	10-year MFS: 71.9% of 6-month ADT vs 78.1% of 2-year ADT	0.773 (0.612–0.975; 0.029)	Toxicity > Grade 3: 14% of 6-month ADT vs 19% of 2-year ADT (P=0.025)
NRG oncology/RTOG 0534 SPPORT11	Phase III/PSA 0.1 ng ml−1 to <2 ng ml−1	6-month ADT + PRT (cohort 2) vs 6-month ADT + PRT + PLNRT (cohort 3)	PRT	FFP	5-year FFP: 70.9% of cohort 1 vs 81.3% of cohort 2 vs 87.4% of cohort 3	P<0.0001	Grade 2 or worse AEs more common in cohort 3 (44%) > cohort 2 (36%; P=0.0034), vs cohort 1 (8%; P<0.0001)

ADT: androgen deprivation therapy; sRT: salvage radiation therapy; PRT: prostate radiation therapy; PLNRT: pelvic lymph node radiation therapy; OS: overall survival; MFS: metastasis-free survival; bRFS: biochemical recurrence-free survival; FFP: failure-free progression; HR: hazard ratio; CI: confidence interval; AEs: adverse events, PSA: prostate-specific antigen; bPFS: biochemical progression-free survival

TWO YEARS OF ADT SHOWS IMPROVEMENT IN SURVIVAL

The RTOG 9601 trial was one of the first pivotal trials that showed that additional ADT improves OS. However, ADT utilized was in the form of bicalutamide 150 mg, a dose that is hardly used today. The study resulted in a 5% actuarial increase in OS at the 12-year survival follow-up (76.3% in the bicalutamide arm versus 71.3% in the placebo group; hazard ratio [HR] for death of 0.77; 95% confidence interval [CI]: 0.59–0.99; P = 0.04).3 While Grade 3 or 4 toxicity was generally similar for both arms, gynecomastia was much higher in the bicalutamide monotherapy group at 70% versus 11%, typical of antiandrogen monotherapy effects. Further subgroup analysis regarding benefit based on pre-sRT PSA level showed that patients who benefited the most were those who had high pre-sRT PSA of >1.5 ng ml⁻¹, whereas those who had low PSA pre-sRT of <0.6 ng ml⁻¹ did not benefit.4 This runs counter to other datasets that suggest the lower PSA pre-sRT generally yields better results. Perhaps, one reason might be the overall higher risk of cardiac and neurologic morbidity that is seen with RTOG 9601.

RADICALS-HD is another trial that was a part of the RADICALS program with different overlapping groups of patients.5 RADICALS-HD was a phase 3 randomized trial that assigned BCR patients in a 3-way 1:1:1 randomization. The randomization schema included a group who received sRT without ADT, another group with short-course 6 months and the third who received long-course 2 years of ADT. There was an additional 2-way 1:1 randomization between sRT with and without short-course 6-month ADT, and the 6 months versus 2 years of ADT all with sRT.6,7 The long-course 2 years of ADT compared to short-course 6 months revealed improved MFS, which is defined as distant metastasis or death from any cause. The entry eligibility criteria were also fairly modest, with a PSA of <5 ng ml⁻¹. In the 2-year versus 6-month ADT with sRT analyses, 762 patients were randomly assigned to 2 years of ADT. Characteristics included majority (41%) of patients had stage IIIA disease, 56% had node-negative disease, although 9% had node-positive prostate cancer and 64% had positive margins, and high-risk disease was seen in 52% of patients. Early sRT was considered in 57% of patients. The trial underwent statistical revision, especially when subsequent studies showed MFS was a viable surrogate end point.8 The results was positive in showing improvement in MFS in the 2-year ADT at a HR of 0.773 (95% CI: 0.612–0.975; P = 0.029) with a 10-year MFS rate of 71.9% in the 6-month group versus 78.1% in the 2-year group. However, no OS benefit from 2 years of ADT was seen. Toxicity reported as Grade 3 or higher was seen in 14% of patients in the 6-month ADT arm compared to 19% in the 2-year ADT group, with similar rates of hematuria and urethral stricture.

SIX MONTHS OF ADT WITH SALVAGE RADIATION THERAPY SHOWED IMPROVEMENT IN PSA OR CLINICAL PROGRESSION BUT NOT SURVIVAL

There were several studies that showed adding 6 months of ADT to sRT improved PSA progression-free survival (PFS). GETUG AFU 16 was a phase III randomized trial that assigned 743 patients with a PSA of ≥0.2 ng ml⁻¹ and <2 ng ml⁻¹ to either 6 months of goserelin with sRT versus sRT alone.9 Results showed better 10-year PFS at 64% (95% CI: 58%–69%) for patients treated with sRT with ADT versus 49% (95% CI: 43%–54%) for patients treated with sRT alone (HR: 0.54, 95% CI: 0.43–0.68; stratified log-rank test P < 0.0001).10 In addition, RADICALS-HD enrolled patients who received short-course 6-month ADT (n = 743 patients) with sRT versus no ADT (n = 737 patients) with sRT.6 The results revealed no difference in the primary endpoints of MFS with a HR of 0.886 (P = 0.35) or in the 10-year MFS at 79.2% in the no-ADT group compared to 80.4% in the 6-month ADT, or in the freedom from distant metastasis with a HR of 0.816 (95% CI: 0.579–1.150; P = 0.24). There was no additional impact from the pre-sRT PSA or comorbidity scores on the findings. However, there was improvement in the clinical PFS with an HR of 0.544 (95% CI: 0.433–0.684; P ≥ 0.0001). This is consistent with the results of the GETUG AFU 16 study. Similarly, RTOG 0534 SPPORT evaluated the question of prostate bed RT (PBRT) alone (group 1) versus PBRT with 6-month ADT (group 2) compared to PBRT with pelvic lymph node radiotherapy (PLNRT) with short-course 6-month ADT (group 3).11,12 Eligible patients had rising PSA of ≥0.1 ng ml⁻¹ to <2.0 ng ml⁻¹ with findings showed freedom from progression (defined as BCR of PSA ≥2 ng ml⁻¹ over PSA nadir) was exceeded when group 1 (no ADT arm) was compared with group 3 (which was both 6-month ADT + PBRT + PLNRT with an overall difference of 17.9%, P < 0.0001), although there was no significant difference between group 2 and 3 (both had 6-month ADT with the difference being the delivery of PBRT only in group 2 vs both PBRT and PLNRT in group 3). These trials taken together established the role of short-course 6-month ADT in addition to sRT with resultant improvement in PSA or clinical progression-free survival, but not overall survival.

THE ROLE OF ARPI AS INTENSIFICATION STRATEGY IN ADDITION TO ADT WITH SRT

Given the promising results of the use of ARPI in metastatic disease, efforts to use it in lieu of ADT or in addition to ADT with sRT have been investigated. FORMULA-509 evaluated the role of intensification of ADT with additional abiraterone acetate and prednisone with apalutamide compared to bicalutamide along with RT.13 The primary end point was 3-year PFS which was technically not met since it did not cross the pre-specified boundary for statistical significance, although the numerical 3-year MFS was in favor of the ADT/AAP/apalutamide arm at 90.6% compared to 87.2% for the ADT + bicalutamide arm. Another phase II trial that evaluated additional intensification with abiraterone acetate with ADT and PBRT was the CARLHA trial that showed adding abiraterone and ADT to PBRT improved the primary end point of 3-year BCR-free survival (RFS) at 81.5% (95% CI: 66.4%–90.3%).14 There are ongoing phase III trials that would evaluate the use of enzalutamide in this sRT with ADT space such as the STEEL trial (clinicaltrials.gov: NCT03809000) or the STARTAR trial (clinicaltrials.gov: NCT03311555). While enzalutamide alone without ADT in combination with RT has been evaluated in the SALV-ENZA trial,15 this is a smaller trial that did not have ADT with sRT as a comparator arm though failure-free progression was favoring the enzalutamide arm over placebo (HR: 0.42; 95% CI: 0.19–0.92; P = 0.031). Regardless, there is still insufficient data to support additional ARPI to shorten the duration of ADT.

HOW TO CHOOSE DURATION OF ADT TREATMENT

The decision regarding ADT duration hinges upon factors beyond a high pre-sRT PSA level. Almost all of these studies included patients whose PSA ranged from 0.1 ng ml⁻¹ to < 2 ng ml⁻¹. However, additional risk factors to consider in adding ADT include presence of multiple adverse pathologic risk factors such as pT3 disease (extracapsular extension or seminal vesicle invasion), lack of positive margins, higher Gleason scores 8, 9, or 10, National Comprehensive Cancer Network (NCCN) high-risk or very high-risk features, short PSA doubling times, and high pre-sRT PSA. These are all potential high-risk factors that perhaps warrant consideration for use of 2-year ADT with sRT, especially since data regarding 2 years of ADT yields improvement in MFS which is not seen with 6 months of ADT. On the other hand, 6 months of ADT may be appropriate for patients with lower risk and absence of adverse pathologic risk factors such as low pre-sRT, slow PSA doubling time (PSADT), intermediate-risk or low-risk NCCN features, positive surgical margins, prolonged time of recurrence from RP, and lack of pT3 disease upon RP. Furthermore, patients who have true localized recurrence, positive surgical margins, protracted PSADT, very low pre-sRT PSA level,4 low Decipher scores,16,17 or the infirm with multiple comorbidities might even be spared from the adverse effects of ADT if they are not expected to benefit from even short-course ADT but would still be appropriate for curative intent sRT alone.

Toxicity from ADT is just an important consideration since Grade 3 or higher toxicity was expectantly higher in those who received long-term ADT at 19% compared to those who were subjected to 6 months of ADT at 14% in the RADICALS-HD comparison (P = 0.025).7 On the other hand, the RTOG 9601 trial revealed an increase in toxicity that led to Grade 3 to 5 events with a preponderance of cardiac and neurologic effects and an increase in the noncancer related deaths in those who were assigned to antiandrogen with sRT,4 suggesting toxicity from higher doses of antiandrogen. Therefore, recognizing toxicity from ADT is of paramount importance to determine the incremental benefits or harms of adding ADT and even ARPI to sRT.

COMPETING INTERESTS

JBAC declares no competing interest specific to this manuscript. JBAC receives Speaker Fees from BMS, EMD Serono, Pfizer, and Merck, and received advisory fees from Pfizer, EMD Serono, AZD, and Merck.