The baseline characteristics of the entire cohort are summarized in Table 1. The median PSA level at the time of mCRPC diagnosis was 16.8 ng/mL (range 0.2–3411.0 ng/mL). Seventy-nine patients (53.0%) had a Gleason score of 9 or 10. Synchronous metastasis was present in 127 patients (85.2%). Fifty-eight patients (38.9%) had oligometastasis at the time of mCRPC diagnosis. The number of patients allocated to good, intermediate and poor prognostic groups was 65 (43.6%), 75 (50.3%) and 9 (6.0%), respectively. Fifty-four patients received docetaxel (65–75 mg/m2) every 3 weeks, with 26 cases (17.4%) before abiraterone treatment. Forty patients (26.8%) underwent cRT, 30 (75.0%) of whom were irradiated before abiraterone failure. The remaining 10 patients (25.0%) received cRT with concurrent secondary hormone therapies after abiraterone failure.
Table 1
Baseline characteristics of the entire cohort (N = 149)
Characteristics | No. (%) |
Age, median (range), years | 68 (45–86) |
PSA at mCRPC | |
≤ 20 ng/mL | 80 (53.7) |
> 20 ng/mL | 69 (46.3) |
Gleason score | |
≤ 8 | 70 (47.0) |
9–10 | 79 (53.0) |
ECOG | |
0–1 | 105 (70.5) |
> 1 | 44 (29.5) |
Prognostic index | |
Good | 65 (43.6) |
Intermediate | 75 (50.3) |
Poor | 9 (6.0) |
Synchronous metastasis | 127 (85.2) |
Oligometastasis | 58 (38.9) |
Chemo-naïve | 123 (82.6) |
PSA prostate-specific antigen, mCRPC metastatic castration-resistant prostate cancer, Chemo-naïve chemotherapy-naïve upon abiraterone treatment |
Thirty patients (20.1%) were in the AbiRT group, whereas 119 patients (79.9%) were in the non-AbiRT group. Compared with the non-AbiRT group, patients in the AbiRT group were more likely to have oligometastasis (P = 0.002), with a lower PSA level at mCRPC diagnosis (P = 0.005). Other baseline characteristics including age, Gleason score, synchronous metastasis, chemotherapy-naïve upon abiraterone treatment and prognostic index were similar (Table 2).
Table 2
Comparison of baseline characteristics in the unmatched and the matched data
Characteristics | Unmatched data | | Matched data |
No. (%) | | No. (%) |
AbiRT (N = 30) | Non-AbiRT (N = 119) | P | | AbiRT (N = 30) | Non-AbiRT (N = 58) | P |
Age, years | | | 0.224 | | | | 0.440 |
≤ 65 | 15 (50.0) | 45 (37.8) | | | 15 (50.0) | 24 (41.3) | |
> 65 | 15 (50.0) | 74 (62.2) | | | 15 (50.0) | 34 (58.6) | |
PSA at mCRPC | | | 0.005 | | | | 0.667 |
≤ 20 ng/mL | 23 (76.6) | 57 (47.9) | | | 23 (76.6) | 42 (72.4) | |
> 20 ng/mL | 7 (23.3) | 62 (52.1) | | | 7 (23.3) | 16 (27.6) | |
Gleason score | | | 0.435 | | | | 0.886 |
≤ 8 | 16 (53.3) | 54 (45.4) | | | 16 (53.3) | 30 (51.7) | |
9–10 | 14 (46.6) | 65 (54.6) | | | 14 (46.6) | 28 (48.3) | |
Prognostic index | | | 0.107 | | | | 0.227 |
Good | 17 (56.6) | 48 (40.3) | | | 17 (56.6) | 25 (43.1) | |
Intermediate/Poor | 13 (43.3) | 71 (59.7) | | | 13 (43.3) | 33 (56.9) | |
Oligometastasis | 19 (63.3) | 39 (32.8) | 0.002 | | 19 (63.3) | 36 (62.1) | 0.908 |
Synchronous metastasis | 24 (80.0) | 103 (86.6) | 0.538 | | 24 (80.0) | 48 (82.8) | 0.750 |
Chemo-naïve | 28 (93.3) | 95 (79.8) | 0.082 | | 28 (93.3) | 53 (91.4) | 1.000 |
PSA prostate-specific antigen, mCRPC metastatic castration-resistant prostate cancer, Chemo-naïve chemotherapy-naïve upon abiraterone treatment |
At a median follow-up of 23.5 months, 54 patients (36.2%) died. Seven patients (4.7%) were lost to follow-up. The median OS of the entire cohort was 38.4 months. The median OS of patients undergoing cRT was not reached, compared with 31.4 months in patients who did not have cRT (P = 0.001). The 2-year OS rates of patients managed by AbiRT, cRT after abiraterone failure, and no cRT was 89.5%, 72.0% and 72.0%, respectively (P = 0.001) (Fig. 2). The median PFS following radiotherapy in the AbiRT group was 12.2 months, and 23 (76.6%) patients had a PSA response after radiotherapy. The median OS of the AbiRT group was not reached whereas, in the non-AbiRT group, the median OS was 31.8 months (P = 0.000). The 2-year OS rates of the AbiRT group and non-AbiRT group were 89.5% and 73.5%, respectively (P = 0.000) (Fig. 3). Upon univariate analysis, AbiRT, oligometastasis, intermediate/poor group according to the prognostic index, PSA > 20 ng/mL and chemotherapy-naïve upon abiraterone treatment were significant prognostic factors for OS (Table 3). Upon multivariate analysis, the AbiRT group [hazard ratio (HR), 0.17; 95% confidence interval (CI), 0.05–0.58; P = 0.004] and intermediate/poor grouping for the prognostic index (HR 2.71; 95% CI, 1.37–5.35; P = 0.004) were significant prognostic factors (Table 3).
Table 3
Univariate and multivariable analyses of factors predictive of overall survival
Variables | Univariate analysis | Multivariate analysis |
| HR (95% CI) | P | HR (95% CI) | P |
AbiRT | | | | |
Yes vs No | 0.15 (0.05, 0.48) | 0.001 | 0.17 (0.05, 0.58) | 0.004 |
Oligometastasis | | | | |
Yes vs No | 0.52 (0.30, 0.93) | 0.028 | 0.75 (0.39, 1.44) | 0.387 |
PSA > 20 ng/mL | | | | |
Yes vs No | 2.16 (1.23, 3.79) | 0.007 | 1.66 (0.93, 2.96) | 0.089 |
Prognostic index | | | | |
Interm/poor vs favorable | 3.11 (1.64, 5.92) | 0.001 | 2.71 (1.37, 5.35) | 0.004 |
Chemo-naïve at Abi | | | | |
Yes vs No | 0.45 (0.25, 0.81) | 0.008 | 0.81 (0.41, 1.61) | 0.548 |
PSA prostate-specific antigen, Interm intermediate, Chemo-naïve chemotherapy-naïve upon abiraterone treatment, HR hazard ratio, CI confidential interval |
After propensity score matching with a caliper of 0.21, 30 patients in the AbiRT group were matched to 58 patients in the non-AbiRT group. The difference between the baseline characteristics was not significant after matching (Table 2). The survival advantage of AbiRT remained significant. The median OS was not reached in the AbiRT group, compared with 44.0 months in the non-AbiRT group (P = 0.009). The 2-year OS of the AbiRT group and non-AbiRT group was 89.5% and 79.3%, respectively (Fig. 3).
The AbiRT group was associated with improved OS in the subgroups of age ≤ 65 years old, PSA ≤ 20 ng/mL and chemotherapy-naïve upon abiraterone treatment (Fig. 4). The benefit of AbiRT was not evaluated in patients with poor prognosis according to the prognostic index given that there were only nine patients in this subgroup. In the subgroup with an intermediate prognosis according to the prognostic index, application of AbiRT was associated with a reduction of death of 85% (HR, 0.15; 95%CI, 0.03–0.63; P = 0.010); this advantage was not observed in patients with a good prognosis (0.23; 0.03–1.79; 0.161).
Radiotherapy was well tolerated, with no grade 3 or higher toxicities reported. Acute side effects included gastrointestinal (GI) toxicity (9 patients), genitourinary (GU) toxicity (11 patients), thrombocytopenia (1 patient), neutropenia (1 patient) and hypokalemia (1 patient). Late GI and GU toxicity were observed in 7 and 6 patients, respectively. Two patients developed vertebral compression fractures, and neither of them were symptomatic.