Purpose:

Outcomes for resistant metastatic castration-resistant prostate cancer (CRPC) are poor. Stereotactic ablative radiotherapy (SABR) induces antitumor immunity in clinical and preclinical studies, but immunologic biomarkers are lacking.

Patients and Methods:

Eighty-nine patients with oligometastatic CRPC were identified by 11C-Choline-PET (Choline-PET) from August 2016 to December 2019 and treated with SABR. Prespecified coprimary endpoints were 2-year overall survival (OS) and PSA progression. Secondary endpoints included 2-year SABR-treated local failure and 6-month adverse events. Correlative studies included peripheral blood T-cell subpopulations before and after SABR.

Results:

128 lesions in 89 patients were included in this analysis. Median OS was 29.3 months, and 1- and 2-year OS were 96% and 80%, respectively. PSA PFS was 40% at 1 year and 21% at 2 years. Local PFS was 84.4% and 75.3% at 1 and 2 years, respectively, and no grade ≥3 AEs were observed. Baseline high levels of tumor-reactive T cells (TTR; CD8+CD11ahigh) predicted superior local, PSA, and distant PFS. Baseline high levels of effector memory T cells (TEM; CCR7CD45RA) were associated with improved PSA PFS. An increase in TTR at day 14 from baseline was associated with superior OS.

Conclusions:

This is the first comprehensive effector T-cell immunophenotype analysis in a phase II trial before and after SABR in CRPC. Results are favorable and support the incorporation of immune-based markers in the design of future randomized trials in patients with oligometastatic CRPC treated with SABR.

Translational Relevance

Although metastasis-directed stereotactic ablative radiotherapy (SABR) benefits 10%–20% of oligometastatic patients, biomarkers that identify which patients will benefit from SABR are lacking. In this prospective trial, patients with oligometastatic CRPC identified by 11C-Choline-PET (Choline-PET) were treated with SABR and antitumor immune parameters were measured before and after radiotherapy from peripheral blood. Significantly, high levels of tumor-reactive effector T cells (TTR) at baseline were associated with improved PSA, local, and distant progression-free survival, and an increase in TTR after SABR was associated with improved overall survival. Baseline TTR can be an immunologic biomarker endpoint in future randomized trial design to identify patients with oligometastatic prostate cancer who will benefit from SABR and help unlock the abscopal potential of radiotherapy in treatment-refractory prostate cancer.

Clinical outcomes for patients with metastatic castration-resistant prostate cancer (CRPC) whose disease is refractory to second-generation antiandrogens and chemotherapy are poor with median overall survival (OS) of 11 to 13.6 months (1). This highlights an urgent need for improved therapeutic strategies. In refractory CRPC, systemic agents are offered with noncurative intent. Similarly, clinicians may also offer localized therapies like radiotherapy; however, most reserve these as purely palliative measures for symptomatic metastases. The extent to which patients with metastatic disease benefit from these localized therapies is an area of active inquiry.

The term “oligometastasis” was coined by Hellman and Weichselbaum (2), and the rationale for aggressive metastasis-directed therapy in this patient population stems from surgical experience; metastasectomies for various solid tumors have demonstrated a 5-year survival rate of 25%–45% (3–5). Although differing theories explain the oligometastatic state, the main premise is that the removal of tumor clones that give rise to metastases will reduce the risk of secondary metastases and thus improve survival. The clinical benefit of stereotactic ablative radiotherapy (SABR) in oligometastatic castration-sensitive prostate cancer (CSPC) is supported by results of recent phase II randomized trials STOMP and ORIOLE (5, 6).

A major obstacle in obtaining favorable outcomes in patients with oligometastatic disease is the identification and selection of patients with a truly limited number of metastases. 11C-Choline-PET (Choline-PET) is more sensitive in detecting pelvic lymph node and distant metastases than traditional imaging modalities with validated sensitivity, specificity, positive predictive value, and negative predictive value at 85%–100%, 76%–96%, 76%–91%, and 81%–100%, respectively (7–12). These appear to be similar to other advanced imaging techniques (13). Therefore, we hypothesized that SABR in choline-PET-identified oligometastatic CRPC may improve outcomes. A goal of this study is to identify blood-based biomarkers that will select patients who benefit from SABR in this setting.

A second major obstacle is the subclinical microscopic disease burden that resists systemic therapy and drives distant failures following SABR. While immune checkpoint inhibition has revolutionized treatment of many cancers, the utility of PD-1/PD-L1 axis inhibition in prostate cancer has been limited (14). Research in engaging antitumor immunity against CRPC is ongoing. A highly promising lead in this area has been the use of tumor-directed radiotherapy to boost immune response (15, 16). Metastasis-directed SABR offers excellent local control and can elicit immune responses in advanced prostate cancer (4, 6–8). We previously demonstrated that SABR and anti-PD-1 therapy induced antigen-specific abscopal responses in a preclinical oligometastatic cancer model, suggesting that such anticancer immune responses may not be limited to certain tumor histologies or host genetic factors (17). These antigen-specific responses are reflected in CD11a-high CD8+ T cells (18). We further demonstrated that tumor-reactive effector T cells predict superior survival in patients with melanoma and are important in driving the abscopal response in preclinical models (17–19). However, clinical data supporting SABR-induced anti–prostate cancer immunity is lacking.

Herein, we report outcomes of our prospective clinical trial that evaluated SABR in Choline-PET–identified oligometastatic CRPC and the induction of anti–prostate cancer immunity based on our prior translational studies (18–20).

Patient population

This study was performed in accordance with the Declaration of Helsinki and approved by the Mayo Clinic Institutional Review Board (IRB). The study was registered with ClinicalTrials.gov (NCT02816983). Informed written consent was obtained from each subject. From August 2016 to December 2019, 89 patients with oligometastatic CRPC were prospectively enrolled on this single-arm, single institution, IRB-approved phase II trial (NCT02816983). Eligibility criteria included histologically confirmed prostate cancer, ≤3 lesions identified on Choline-PET/CT performed as described previously (21, 22), castrate levels of testosterone (< 50 ng/dL) on androgen deprivation therapy (ADT), ECOG performance status score of 0–2, >6 months of life expectancy, and consent for blood draws. Coprimary endpoints were 2-year OS and PSA progression (Prostate Cancer Working Group criteria; ref. 23). Secondary endpoints included 2-year SABR-treated local failure (LF, defined radiographically) and 6-month rates of adverse events (AE) per Common Terminology Criteria for Adverse Events (CTCAE) v4.0. Peripheral blood samples were collected for immunologic biomarker correlative analyses as described previously (20, 24).

Statistical analysis

Coprimary endpoints were OS and PSA progression while the secondary endpoints included local failure (LF) of SABR-treated lesion(s) and distant failure (DF). Hussain and colleagues demonstrated that PSA progression in metastatic CRPC predicts OS, and the median PSA PFS was 8.1 months for the SWOG 99–16 cohort using the PCWG 2008 PSA progression definition (25). Overall survival was 17 months. This translates to a one-year PSA PFS of 35.8% assuming an exponential distribution. Assuming α 0.05, our cohort of 89 patients was powered at 80% to detect a statistically significant difference in survival. Of these, 9 were treated with conventional radiotherapy due to insurance denial of SABR while the remaining 80 received SABR treatment. Descriptive statistics are reported as median and interquartile range (IQR) for continuous values and reporting of median follow-up and median survival. For discrete variables, the number and percentage are reported. Follow-up visits were routinely performed every 3–4 months in person (or telephone with mail-in PSA) at physician's discretion. The minimum time from SABR to follow-up imaging was at least 3 months and any increase in PSA triggered imaging with choline PET. LF and DF were, respectively, defined as choline uptake at the site of prior SABR-treated or elsewhere. A positive 11C-choline PET/CT was defined as a PET/CT study that identified a tracer-avid lesion (focal tracer uptake greater than the surrounding blood pool). For LF, an increase in SUVmax value after SABR, and when available, expansion of bony metastasis on CT or MRI was used. For new spine or pelvic metastases, MR was performed to verify choline findings. For extremity bone, PET finding was confirmed on planning CT. PSA progression was defined per Prostate Cancer Working Group criteria (25% increase between PSA measured at any of the six follow-up time points and nadir of 2 ng/mL or greater; ref. 23).

We verified patient survival from the third-party agent Accurint in addition to the medical record. For all other outcomes, data were collected up to and censored at the most recent follow-up visit. Kaplan–Meier estimates were used to estimate survival curves and a univariate Cox model was used to determine the association of patient and disease variables with the risk of each outcome of interest. Expected 2-year overall survival was estimated at 30%–35% for the powering of this study. The alpha-level for statistical significance was prespecified at 0.05. OS and recurrence-free survival (biochemical, local, and distant) estimates are reported with 95% confidence intervals. The alpha-level for statistical significance was prespecified at 0.05.

Immunologic biomarkers were measured as continuous variables for correlative studies and converted to a categorical variable to classify patients as high versus low risk. We used the Contal and O'Quigley method (26) based on the log rank test statistic to estimate the cut-off value needed to categorize each immune cell quantity. The best cut-off value for each survival outcome was calculated on the basis of baseline values. These cut-off values were then used at all other time points. Cutoffs were 7% for tumor-reactive T cells (TTR), 44% for PD-1–positive TTR, and 24% for TEM.

Imaging technique

11C-choline PET/CT was performed on an integrated PET/CT scanner (Discovery LS, RX, 690, or 710, GE Healthcare) with the use of 11C-choline produced at our on-site cyclotron facility as described previously (17, 19). A CT scout scan was performed to define the body axial range to be imaged. Next, each patient received a single-dose, intravenous bolus injection of 555 to 740 MBq 11C-choline (half-life 20.4 minutes). Low-dose helical CT images were then obtained with the patient doing shallow breathing for attenuation correction and anatomic localization (detector row configuration, 16 × 0.625 mm; pitch, 1.75; gantry rotation time, 0.5 seconds; slice thickness, 3.75 mm; 140 kVp; and range 60–120 mA with the use of automatic current modulation) followed by PET acquisition initiated at approximately 5 minutes after injection. PET images were acquired from mid-thigh to the orbits (in three dimensions, with a 128 × 128 matrix and at a rate of 3-to 4 minutes per bed position depending on body mass index). PET images were reconstructed with a three-dimensional ordered-subsets expectation maximization algorithm (28 subsets, two iterations).

SABR technique

Patient immobilization, SABR planning, and delivery were performed as described previously (2, 10). In brief, SABR was delivered in 1 to 5 fractions (range, 20–50 Gy total) in accordance with AAPM Task Group 101 recommendations (27). Modal dose and fractionation were 20 Gy and 1 fraction, respectively.

We highlight that our SABR planning process incorporates diagnostic Choline-PET image fusion with planning CT to delineate target volumes. MRI acquired in the radiotherapy treatment position was obtained for spine metastases, which confirmed target volumes and helped to define the spinal cord and cauda equina. For spine lesions, clinical target volume (CTV) was defined per RTOG 0631 (28). Five millimeter planning target volume (PTV) expansion was used where appropriate. In cases of spine lesions, 2–3 mm was used.

Immunologic studies

We collected peripheral blood mononuclear cells (PBMC) at four time points: at baseline and on days 1, 7, and 14 post-SABR. We performed two-color staining for CD11a and CD8 to monitor the frequency of TTR (CD8+CD11ahigh; refs. 18, 20, 24). Markers CCR7 and CD45RA defined predetermined subsets of CD8+ T cells assigned to one of four subgroups based upon the intensity of biomarker staining: naïve (CCR7+CD45RA+), central memory (TCM; CCR7+CD45RA), effector/effector memory (TEM; CCR7CD45RA), and terminally differentiated (CCR7CD45RA+). We analyzed PBMCs both at baseline and at defined time points after SABR as described above using flow cytometry. We analyzed between 10,000 and 100,000 peripheral blood T cells per patient according to cell recovery.

Flow analysis of human T cells isolated from peripheral blood

We identified PBMC subpopulations with the following antibody panel: CD8-PE-Cy7 (BD Pharmingen, clone RPA-T8, catalog no. 304006), CD11a-APC (BioLegend, clone HI111, catalog no. 301212), PD-1 FITC (BioLegend, clone EH12.2H7, catalog no. 32990), Ki-67-BV421 (BD Biosciences, clone B56, catalog no. 562899), CCR7-BV650 (BD Biosciences, clone 2-L1-A, catalog no. 566756), and CD45RA (BD Biosciences, clone HI100, catalog no. 563031). We stained surface markers prior to intracellular markers. Flow cytometry was performed at the institutional flow cytometry core facility on a FACS Canto II (BD Biosciences) and data were collected on a CytoFLEX LX (Beckman Coulter). We analyzed flow cytometry data with FlowJo 10.4 (Tree Star).

Patient and treatment characteristics

Baseline patient characteristics are presented in Table 1. Median follow-up was 23 months (IQR, 11–25), and median age was 71 years (range, 51–84; IQR, 64–75). Eighty-nine patients with 128 total metastatic lesions treated were included in the analysis. Fifty-eight patients (65%) had one lesion, 23 (26%) had two lesions, and eight (9%) had three lesions. Nine of the 89 subjects received conventional radiotherapy as noted in the methods (8 Gy in 1 fraction). Median PSA before SABR was 0.7 ng/mL (IQR, 0.25–3.1), and 64% had PSA < 2 ng/mL. Prior to SABR (n = 80), 61%, 44%, and 19% had disease progression while receiving chemotherapy, abiraterone, or enzalutamide, respectively. Significantly, 38% of the cohort had progressed on both chemotherapy and at least one second-generation antiandrogens. Seventy-one percent of patients had received prior SABR to the prostate or prostate fossa. Patients were maintained on the same systemic therapy before and after SABR until clinical failure.

Table 1.

Patient and treatment characteristics.

‘Table of baseline Demographic and Treatment variables’
Total (N = 89)
Age 
 Mean (SD) 69.7 (7.0) 
 Median 70.8 
 Q1, Q3 63.8, 74.7 
 Range (51.0–83.9) 
Race/Ethnicity 
 White 83 (93.3%) 
 Asian 1 (1.1%) 
 Unknown or not reported 4 (4.5%) 
 Other 1 (1.1%) 
T score at initial diagnosis 
 Missing 
 T1c 3 (3.6%) 
 T2a 7 (8.4%) 
 T2b 3 (3.6%) 
 T2c 15 (18.1%) 
 T3a 18 (21.7%) 
 T3b 34 (41.0%) 
 T4 3 (3.6%) 
N score at time of diagnosis 
 Missing 
 NX 7 (8.0%) 
 N0 56 (64.4%) 
 N1 23 (26.4%) 
 N2 1 (1.1%) 
M score at time of diagnosis 
 Missing 
 MX 21 (25.9%) 
 M0 54 (66.7%) 
 M1 6 (7.3%) 
SABR Dose (Gy) (all 128 sites treated) 
 N 128 
 Mean (SD) 27.45 (11.3) 
 Median 23.0 
 Q1, Q3 20.0, 39.0 
 Range (8.0–50.0) 
Number of fractions (all 128 sites treated) 
 N 128 
 Mean (SD) 2.1 (1.3) 
 Median 1.0 
 Q1, Q3 1.0, 3.0 
 Range (1.0–5.0) 
Number of sites treated 
 1 58 (65.2%) 
 2 23 (25.8%) 
 3 8 (9.0%) 
Sites treated 
 Bone only: spine 41 (46.1%) 
 Bone only: nonspine 30 (33.7%) 
 Lymph nodes only 5 (5.6%) 
 Bone and lymph nodes 12 (13.5%) 
 Other 1 (1.1%) 
ECOG Performance score 
 0 74 (83.1%) 
 1 15 (16.9%) 
Prior radiation 77 (85%) 
 Primary prostate 23 (25.8%) 
 Prostate bed/pelvic lymph nodes 22 (24.7%) 
 Metastatic disease 17 (19.1%) 
 Primary prostate and metastatic disease 10 (11.2%) 
 Prostate bed and metastatic disease 8 (9.0%) 
Prior therapy 
 Chemotherapy/docetaxel 54 (60.7%) 
 Abiraterone 39 (43.8%) 
 Enzalutamide 17 (19.1%) 
 Both chemotherapy + abiraterone 30 (33.7%) 
 Both chemotherapy + enzalutamide 10 (11.2%) 
Androgen deprivation method 
 Leuprolide 87 (97.8%) 
 Goserelin 2 (2.2%) 
 Orchiectomy 2 (2.2%) 
Prostatectomy (including salvage) 
 No 13 (14.9%) 
 Yes 74 (85.1%) 
 Missing 
‘Table of baseline Demographic and Treatment variables’
Total (N = 89)
Age 
 Mean (SD) 69.7 (7.0) 
 Median 70.8 
 Q1, Q3 63.8, 74.7 
 Range (51.0–83.9) 
Race/Ethnicity 
 White 83 (93.3%) 
 Asian 1 (1.1%) 
 Unknown or not reported 4 (4.5%) 
 Other 1 (1.1%) 
T score at initial diagnosis 
 Missing 
 T1c 3 (3.6%) 
 T2a 7 (8.4%) 
 T2b 3 (3.6%) 
 T2c 15 (18.1%) 
 T3a 18 (21.7%) 
 T3b 34 (41.0%) 
 T4 3 (3.6%) 
N score at time of diagnosis 
 Missing 
 NX 7 (8.0%) 
 N0 56 (64.4%) 
 N1 23 (26.4%) 
 N2 1 (1.1%) 
M score at time of diagnosis 
 Missing 
 MX 21 (25.9%) 
 M0 54 (66.7%) 
 M1 6 (7.3%) 
SABR Dose (Gy) (all 128 sites treated) 
 N 128 
 Mean (SD) 27.45 (11.3) 
 Median 23.0 
 Q1, Q3 20.0, 39.0 
 Range (8.0–50.0) 
Number of fractions (all 128 sites treated) 
 N 128 
 Mean (SD) 2.1 (1.3) 
 Median 1.0 
 Q1, Q3 1.0, 3.0 
 Range (1.0–5.0) 
Number of sites treated 
 1 58 (65.2%) 
 2 23 (25.8%) 
 3 8 (9.0%) 
Sites treated 
 Bone only: spine 41 (46.1%) 
 Bone only: nonspine 30 (33.7%) 
 Lymph nodes only 5 (5.6%) 
 Bone and lymph nodes 12 (13.5%) 
 Other 1 (1.1%) 
ECOG Performance score 
 0 74 (83.1%) 
 1 15 (16.9%) 
Prior radiation 77 (85%) 
 Primary prostate 23 (25.8%) 
 Prostate bed/pelvic lymph nodes 22 (24.7%) 
 Metastatic disease 17 (19.1%) 
 Primary prostate and metastatic disease 10 (11.2%) 
 Prostate bed and metastatic disease 8 (9.0%) 
Prior therapy 
 Chemotherapy/docetaxel 54 (60.7%) 
 Abiraterone 39 (43.8%) 
 Enzalutamide 17 (19.1%) 
 Both chemotherapy + abiraterone 30 (33.7%) 
 Both chemotherapy + enzalutamide 10 (11.2%) 
Androgen deprivation method 
 Leuprolide 87 (97.8%) 
 Goserelin 2 (2.2%) 
 Orchiectomy 2 (2.2%) 
Prostatectomy (including salvage) 
 No 13 (14.9%) 
 Yes 74 (85.1%) 
 Missing 

Disease outcomes

OS

Eighteen deaths were observed with median OS of 29.3 months. OS at 1 and 2 years was 95.9% [95% confidence interval (CI), 91.5%–100.0%] and 80.2% (95% CI, 69.9%–91.9%), respectively (Fig. 1A). No significant difference in OS was noted between patients with one or more than one metastasis (HR, 0.55; 95% CI, 0.16–1.9; P = 0.33). In multivariate analyses, number of metastases did not significantly change outcomes.

Figure 1.

Patient outcomes. A, Overall survival of all patients in the study. B, PSA PFS of all patients in the study. C and D, Local and distant PFS by modality (conventional versus stereotactic radiotherapy).

Figure 1.

Patient outcomes. A, Overall survival of all patients in the study. B, PSA PFS of all patients in the study. C and D, Local and distant PFS by modality (conventional versus stereotactic radiotherapy).

Close modal

PSA progression-free survival

Sixty-three patients experienced PSA progression within 2 years of treatment. Median time to PSA progression for these patients was 9.2 months (95% CI, 7.0–12.4 months). PSA progression-free survival (PFS) was 39.5% at 1 year (95% CI, 30.1%–51.8%) and 20.8% at 2 years (95% CI, 13.1%–33%; Fig. 1B). Median time to PSA progression was 9.5 months for patients treated with SABR and 7.7 months for patients treated with conventional radiotherapy. Seven of the 9 patients in the conventional radiotherapy group experienced PSA progression within the first year. One patient was lost to follow-up at 15 months and the remaining patient progressed at approximately 23 months. PSA PFS at one year was 40.9% (95% CI, 31%–54%) for patients treated with SABR compared with 27.8% (95% CI, 8.9%–86.9%) for patients treated with conventional radiotherapy. At two years, the PSA PFS for patients treated with SABR was estimated at 22.5% (95% CI, 14.3%–35.4%). Conversely, all patients with conventional radiotherapy experienced disease progression.

The number of metastatic sites correlated with PSA progression-free survival (Supplementary Fig. S1A; Supplementary Table S1). Patients with three treated metastatic sites experienced significantly inferior PSA progression-free survival (median 5 months) relative to patients with two metastases (median 10.9 months; HR, 0.33; 95% CI, 0.13–0.83; P = 0.018) or one metastasis (9.7 months; HR, 0.33; 95% CI, 0.14–0.76; P = 0.019).

Local progression of SABR-treated metastases and distant progression

Twenty-one patients experienced LF after radiotherapy. Median time to LF was not reached. Local PFS was 84.4% and 75.3% at 1 and 2 years, respectively (Fig. 1C).

A total of 66 patients experienced DF with a median time to DF of 5.1 months (95% CI, 3.6–6.9). Distant PFS for all patients was 17.6% (95% CI, 10.6–29.5%) and 5.0% (95% CI, 1.7%–14.8%) at 1 and 2 years, respectively. All conventional radiotherapy patients had distant recurrences by 9 months. The 1- and 2-year distant PFS for patients receiving SABR was 19.4% (95% CI, 11.6%–32.2%) and 5.5% (95% CI, 1.9%–16.2%), respectively (Fig. 1D). Distant PFS was not significantly related to the number of metastases detected at baseline (Supplementary Fig. S1B; Supplementary Table S2).

Adverse events

Adverse events at 6 months were not evaluable in 17 patients (20%). Among 72 evaluable patients, no grade ≥ 3 AEs were observed. Grade 2 AEs included pain flare (8.5%), fracture (2.9%), and fatigue (1.4%).

Immunologic parameters and outcomes

We previously validated CD8+ CD11ahigh TTR T-cell populations as significant in anti-tumor immunity (18, 20, 24). To determine whether these cells or other known anti-tumor T cell populations predict patient outcomes, we measured these cells in systemic circulation before and after radiotherapy (see Supplementary Fig. S2).

Baseline TTR levels predicted significantly improved PSA PFS, local PFS, and distant PFS (Fig. 2AC; Supplementary Table S3). Patients with baseline TTR levels above the cutoff experienced prolonged median time to PSA progression compared with patients with levels below the cutoff (9.7 vs. 5.6 months; HR 0.57, 95% CI [0.3–0.99]; P = 0.04). Similarly, patients with baseline TTR above the cutoff did not reach median time to LF while their counterparts achieved a median 23 months local PFS (HR 0.27 [95% CI, 0.09–0.80]; P = 0.01). Interestingly, patients with PD-1+ TTR at baseline above the cutoff experienced shorter median time to LF (9.1 vs. not reached; HR 6.45, 95% CI, 1.68–24.79; P = 0.002).

Figure 2.

High baseline TTR levels predict superior PSA, local, and distant PFS. A, Biochemical progression-free survival. B, Local progression-free survival. C, Distant progression-free survival.

Figure 2.

High baseline TTR levels predict superior PSA, local, and distant PFS. A, Biochemical progression-free survival. B, Local progression-free survival. C, Distant progression-free survival.

Close modal

Median time to DF was also improved in patients with TTR above the cutoff compared with their counterparts with TTR below the cutoff (6.4 vs. 3.5 months; HR 0.44; 95% CI, 0.24–0.83; P = 0.009).

Next, we correlated clinical outcomes with baseline T-cell subgroups: naïve (TN, CCR7+CD45RA+), central memory (TCM, CCR7+CD45RA), effector memory (TEM, CCR7CD45RA), and terminally differentiated (TTD, CCR7CD45RA+). Patients with TEM at baseline above the cutoff achieved prolonged median time to PSA progression (10.5 vs. 6.2 months; HR, 0.46; 95% CI, 0.22–0.96; P = 0.03).

We next assessed whether CD8+ T-cell population changes after SABR predict clinical outcomes by assessing the percent change of various effector T-cell levels pre- and post-SABR (Fig. 3; Supplementary Fig. S3). Patients with increased TTR at day 14 experienced improved OS compared with those with decreased TTR at day 14 (median OS, 33.3 months vs. 28.2 months; HR, 6.21; P = 0.04). Conversely, there was a trend for worse survival with an increase in TCM on day 1 (median OS 32.7 vs. 29.3 months; HR 0.18; P = 0.08). Increasing TCM consistently predicted inferior local control (median time to LF not reached vs. 15.8 months; P = 0.053). Increasing Ki67 in TTR (P = 0.042) and TEM (P = 0.049) populations at days 1 and 7, respectively, predicted improved local control.

Figure 3.

TTR cell changes after radiotherapy predict OS in oligometastatic CRPC. A, TTR cells are defined by expression of both CD8 and CD11a. B, Patients with increasing TTR levels at day 14 after radiotherapy to oligometastases experience superior OS in CRPC.

Figure 3.

TTR cell changes after radiotherapy predict OS in oligometastatic CRPC. A, TTR cells are defined by expression of both CD8 and CD11a. B, Patients with increasing TTR levels at day 14 after radiotherapy to oligometastases experience superior OS in CRPC.

Close modal

Patients with increasing TTR at day 1 achieved superior median time to DF (6.7 vs. 3.6 months; HR 0.49, 95% CI, 0.2–0.84; P = 0.008). In addition, patients with increasing TEM at day 14 achieved superior median time to PSA progression (median not reached vs. 7.9 months; HR, 2.69; 95% CI, 1.03–7.06; P = 0.04).

To our knowledge, this study is both the first prospective PET-informed metastasis-directed SABR trial in oligoprogressive CRPC and the first to correlate response to SABR therapy with immune cell populations. Two-year OS and PSA PFS were 80% and 21%, respectively. SABR was well tolerated with no grade ≥ 3 AEs. Baseline levels of TTR (CD8+CD11ahigh) effector T cells and TEM (CCR7CD45RA) predicted improved PSA and local PFS. Post-SABR increases in TTR and TEM were associated with improved distant and PSA PFS, respectively, and an increase in Ki67 TTR and TEM was associated with improved local control. Conversely, TCM correlated with inferior OS and local control.

Emerging randomized trials have determined a clinical benefit of SABR in patients with oligometastatic disease (5, 6, 29, 30). In CSPC, SABR improved 5-year ADT-free survival in STOMP (31) and median PFS in ORIOLE (5). Oligometastasis-directed SABR is safe and locally effective treatment of metastatic prostate cancer. SABR may delay more toxic systemic therapies as demonstrated in the recent NRG-BR001 phase I trial in a spectrum of solid tumor types (32).

In this study, we selected OS and PSA PFS as co-primary endpoints. Our cohort comprised advanced, CRPC with resistance to multi-systemic therapy. The single-arm design of the trial reflected clinician concern that equipoise between observation and SABR in patients with advanced CRPC was not possible. Despite resistance to or intolerance of prior chemotherapy (61%), abiraterone or enzalutamide (44%), or both chemotherapy and a next generation antiandrogen (38%) prior to SABR, our patients with oligometastatic CRPC experienced median OS and PSA PFS of 29.3 and 9.2 months, respectively. In the CARD trial, OS was 13.6 months and PFS was 4.4 months (1). The CARD trial included patients with similar characteristics, although different imaging modalities were used to define metastatic disease and some non-metastatic disease was included in the CARD trial. Choline-PET-directed SABR with or without novel systemic therapy warrants further investigation.

The precise imaging afforded by Choline-PET offers several advantages. First, highly specific modalities identify patients with metastatic disease in whom additional treatment is effective. Second, highly sensitive modalities select patients with truly oligometastatic disease. Although this selection may inflate survival by excluding patients with more advanced disease previously included in such studies (33), it crucially also rules out patients for whom such interventions are not effective. In this study, the presence of a third metastasis predicted inferior outcomes despite localized therapy. We can infer that Choline-PET successfully selected oligometastatic disease from a cohort whose disease is highly resistant to systemic therapy for a highly promising local treatment modality. Equipoise was not present in our population to ethically conduct a randomized controlled trial with an observational arm.

Regarding SABR-treated metastasis, local PFS of 84.4% and 75.3% at 1 and 2 years were lower than seen in some prior studies. In our previous report of patients with CSPC, local PFS at 2 years was 95% with single SABR fraction dose ≥ 18 Gy (2). In our follow-up study of 201 patients, 1- and 3-year local PFS were 95.5% and 87.1%, respectively (34). In addition, castration resistance in prostate cancer is commonly associated with altered AR expression and signaling, which is associated with broad chemo- and radioresistance as previously demonstrated (35). These results are not unexpected, as resistance mechanisms in prostate cancers are commonly distal to DNA-damaging small molecules, radiation, and biologics (36).

We examined the impact of SABR and its ability to induce anti–prostate cancer immunity in our CRPC cohort. We have previously identified specific T-cell “states” (TTR and TAC) that have important tumoricidal function following photon radiation. Consistent with the importance of PD-1, we showed that effector CD8+ T cells positive for PD-1 and CD11ahigh are induced after tumor irradiation and are tumor reactive (24). Using this biomarker, we monitored the frequency of prostate cancer reactive CD8+ T cells in the peripheral blood. Cutoffs were retrospective and these observations will require confirmation.

We previously reported that SABR can induce systemic antitumor immune response by promoting peripheral expansion of TTR with an effector phenotype (20, 24). Expansion of these specific TTR is a prerequisite for an active local and systemic antitumor immune response (abscopal effect). Herein, we further underscore the role of SABR in antitumor immunity where expansion of TTR and TEM was associated with improved clinical outcomes. However, high rates of distant failure suggest that other determinants impair SABR-induced antitumor immunity. Radiotherapy upregulates expression of PD-1 on TTR which induce T-cell apoptosis upon engagement with PD-L1. In our study, patients with PD-1+ TTR at baseline above the cutoff experienced shorter median time to LF (9.1 vs. not reached; P = 0.002). We can speculate whether the PD-1/PD-L1 axis blockade along with SABR may rescue the exhausted TTR and change the clinical course. As demonstrated previously, the combination of SABR with PD-1 blockade may prevent T-cell death and facilitate an effective and sustained antitumor immune response (37). In patients with melanoma, circulating tumor-reactive PD-1+ TTR with elevated expression of Bim, a promoter of apoptosis downstream from PD-1, was a predictor of poor survival in patients who did not receive anti–PD-1 therapy; further, an anti–PD-1 therapy-induced reduction of Bim in TTR cells was associated with improved survival (20, 24). There is a striking report of a patient with immunotherapy-resistant melanoma who was treated with local radiation and anti-PD-1 therapy experienced a dramatic response to therapy coupled with decreased Bim in TTR cells (manuscript under review), which highlights the potential interaction between radiation and antimelanoma immunity.

We report a similar dependence on TTR cells for systemic immunity induction following SBRT in oligometastatic prostate cancer. We also have identified actively cytotoxic CD8+ effector T (TAC) cells associated with the abscopal effect that can migrate into tumors and execute tumoricidal functions (20). These data highlight the interplay between radiation and anti-tumor immunity and support our hypothesis that specific T-cell subsets (TTR and TAC) are key mediators of antitumor immunity associated with the abscopal effect. LET optimized proton and carbon ion therapy create more complex DNA damage and greater cell kill than conventional high energy photons. Based on these physical properties, investigations are ongoing whether higher LET radiations may differentially affect local and systemic antitumor immunity.

Beyond quantitating differences in TTR and TAC subsets and their functionality, we examined T effector differentiation states and their proportions following SABR. An increase in central and effector memory cells, which should be seen within one to two weeks, would suggest that radiotherapy was effective in inducing antitumor responses, irrespective of prostate tumor antigen reactivity. Our study suggests that these cells may help determine prognosis as well as direct future treatment modalities to overcome resistance. This may simply reflect the concomitant decrease in TEM in these same patients. The baseline presence of TTR and TEM suggest there are preexisting tumor antigen–primed T cells, and these T cells can be further induced into expansion upon re-activation by tumor antigens that are released during SBRT. While the natural antitumor immune response does lead to some local control benefit, this also shows potential synergy with immune checkpoint blockade therapy.

Conclusion

In this study, we treated 89 patients with oligometastatic CRPC identified by Choline-PET. We observed an estimated 2-year OS of 80% in this heavily treated population after metastasis-directed therapy with no grade ≥ 3 AEs. High levels of tumor reactive (TTR, CD8+CD11ahigh) effector T cells at baseline were associated with improved PSA, local, and distant PFS. Radiotherapy may also indirectly induce T-cell changes through either antigen release or proinflammatory cytokine production. These results are favorable compared to other treatment modalities and support the incorporation of immune-based markers in the design of future clinical trials in patients with oligometastatic CRPC treated with SABR.

V.J. Lowe reports personal fees from Eisai Inc, AVID Radiopharmaceuticals, and Merck Research and grants from General Electric, Siemens Molecular Imaging, and NIH during the conduct of the study as well as grants from AVID Radiopharmaceuticals outside the submitted work. E.J. Tryggestad reports grants from NCI/NIH during the conduct of the study. L.C. Pagliaro reports other support from Merck, Astellas, Pfizer, Exelixis, and Roche/Genentech outside the submitted work. M.J. Iott reports grants from NIH/NCI during the conduct of the study. B.A. Costello reports other support from Exelixis and Clinical Care Options outside the submitted work. G.B. Johnson reports other support from Curium; grants from MedTrace and Blue Earth; and grants and other support from Novartis, Pfizer, Clarity, and ViewPoint outside the submitted work; in addition, G.B. Johnson has a patent for radionuclide imaging and therapy pending. E.D. Kwon reports a patent for B7-H1 issued, licensed, and with royalties paid from BMS. S.S. Park reports grants from NIH/NCI during the conduct of the study as well as grants from MacroGenics and other support from AstraZeneca outside the submitted work; in addition, S.S. Park has a patent for Assessing and Treating Prostate Cancer issued. No disclosures were reported by the other authors.

H. Zhang: Formal analysis, investigation, methodology, writing–original draft, writing–review and editing. J.J. Orme: Data curation, formal analysis, investigation, writing–original draft, project administration, writing–review and editing. F. Abraha: Formal analysis. B.J. Stish: Conceptualization, resources, methodology, writing–review and editing. V.J. Lowe: Funding acquisition, investigation, writing–review and editing. F. Lucien: Investigation, methodology, writing–review and editing. E.J. Tryggestad: Investigation, methodology, writing–review and editing. M.S. Bold: Investigation, methodology, writing–review and editing. L.C. Pagliaro: Investigation, writing–review and editing. C.R. Choo: Investigation, methodology, writing–review and editing. D.H. Brinkmann: Investigation, methodology, writing–review and editing. M.J. Iott: Supervision, methodology, writing–review and editing. B.J. Davis: Investigation, methodology, writing–review and editing. J.F. Quevedo: Investigation, methodology, writing–review and editing. W.S. Harmsen: Data curation, formal analysis, writing–review and editing. B.A. Costello: Investigation, methodology, writing–review and editing. G.B. Johnson: Conceptualization, investigation, writing–review and editing. M.A. Nathan: Conceptualization, investigation, methodology, writing–review and editing. K.R. Olivier: Conceptualization, resources, supervision, funding acquisition, methodology, writing–review and editing. T.M. Pisansky: Conceptualization, supervision, investigation, methodology, writing–review and editing. E.D. Kwon: Conceptualization, resources, supervision, funding acquisition, investigation, methodology, writing–review and editing. H. Dong: Conceptualization, resources, supervision, funding acquisition, methodology, writing–review and editing. S.S. Park: Conceptualization, resources, formal analysis, supervision, funding acquisition, investigation, methodology, writing–original draft, project administration.

We thank Ms. Christina C. Todd (research coordinator), Ms. Brianna N. Tranby (assistance in manuscript preparation), and our patients who participated in this study. This research was supported by NCI R01 CA200551 (to S.S. Park, V.J. Lowe, D.H. Brinkmann, E.D. Kwon, and H. Dong).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
de Wit
R
,
Kramer
G
,
Eymard
J-C
,
de Bono
JS
,
Sternberg
CN
,
Fizazi
K
, et al
CARD: randomized, open-label study of cabazitaxel (CBZ) vs abiraterone (ABI) or enzalutamide (ENZ) in metastatic castration-resistant prostate cancer (mCRPC)
.
Ann Oncol
2019
;
30
:
v882
3
.
2.
Muldermans
JL
,
Romak
LB
,
Kwon
ED
,
Park
SS
,
Olivier
KR
. 
Stereotactic body radiation therapy for oligometastatic prostate cancer
.
Int J Radiat Oncol
2016
;
95
:
696
702
.
3.
Jemal
A
,
Siegel
R
,
Xu
J
,
Ward
E
. 
Cancer statistics, 2010
.
CA Cancer J Clin
2010
;
60
:
277
300
.
4.
Triggiani
L
,
Mazzola
R
,
Magrini
SM
,
Ingrosso
G
,
Borghetti
P
,
Trippa
F
, et al
Metastasis-directed stereotactic radiotherapy for oligoprogressive castration-resistant prostate cancer: a multicenter study
.
World J Urol
2019
;
37
:
2631
7
.
5.
Phillips
R
,
Shi
WY
,
Deek
M
,
Radwan
N
,
Lim
SJ
,
Antonarakis
ES
, et al
Outcomes of observation vs stereotactic ablative radiation for oligometastatic prostate cancer
.
JAMA Oncol
2020
;
6
:
650
.
6.
Ost
P
,
Reynders
D
,
Decaestecker
K
,
Fonteyne
V
,
Lumen
N
,
De Bruycker
A
, et al
Surveillance or metastasis-directed therapy for oligometastatic prostate cancer recurrence: a prospective, randomized, multicenter phase II trial
.
J Clin Oncol
2018
;
36
:
446
53
.
7.
Tran
PT
,
Ost
P
. 
Developments in oligometastatic hormone-sensitive prostate cancer
.
World J Urol
2019
;
37
:
2545
7
.
8.
Demaria
S
,
Ng
B
,
Devitt
ML
,
Babb
JS
,
Kawashima
N
,
Liebes
L
, et al
Ionizing radiation inhibition of distant untreated tumors (abscopal effect) is immune mediated
.
Int J Radiat Oncol
2004
;
58
:
862
70
.
9.
Husarik
DB
,
Miralbell
R
,
Dubs
M
,
John
H
,
Giger
OT
,
Gelet
A
, et al
Evaluation of [18F]-choline PET/CT for staging and restaging of prostate cancer
.
Eur J Nucl Med Mol Imaging
2008
;
35
:
253
63
.
10.
Ahmed
KA
,
Barney
BM
,
Davis
BJ
,
Park
SS
,
Kwon
ED
,
Olivier
KR
. 
Stereotactic body radiation therapy in the treatment of oligometastatic prostate cancer
.
Front Oncol
2013
;
2
:
215
.
11.
de Jong
IJ
,
Pruim
J
,
Elsinga
PH
,
Vaalburg
W
,
Mensink
HJ
. 
Preoperative staging of pelvic lymph nodes in prostate cancer by 11C-choline PET
.
J Nucl Med
2003
;
44
:
331
5
.
12.
Breeuwsma
AJ
,
Pruim
J
,
van den Bergh
ACM
,
Leliveld
AM
,
Nijman
RJM
,
Dierckx
RAJO
, et al
Detection of local, regional, and distant recurrence in patients with PSA relapse after external-beam radiotherapy using 11C-choline positron emission tomography
.
Int J Radiat Oncol
2010
;
77
:
160
4
.
13.
Evans
JD
,
Jethwa
KR
,
Ost
P
,
Williams
S
,
Kwon
ED
,
Lowe
VJ
, et al
Prostate cancer–specific PET radiotracers: a review on the clinical utility in recurrent disease
.
Pract Radiat Oncol
2018
;
8
:
28
39
.
14.
Hansen
AR
,
Massard
C
,
Ott
PA
,
Haas
NB
,
Lopez
JS
,
Ejadi
S
, et al
Pembrolizumab for advanced prostate adenocarcinoma: findings of the KEYNOTE-028 study
.
Ann Oncol
2018
;
29
:
1807
13
.
15.
Lugade
AA
,
Moran
JP
,
Gerber
SA
,
Rose
RC
,
Frelinger
JG
,
Lord
EM
. 
Local radiation therapy of B16 melanoma tumors increases the generation of tumor antigen-specific effector cells that traffic to the tumor
.
J Immunol
2005
;
174
:
7516
23
.
16.
Gulley
JL
,
Arlen
PM
,
Bastian
A
,
Morin
S
,
Marte
J
,
Beetham
P
, et al
Combining a recombinant cancer vaccine with standard definitive radiotherapy in patients with localized prostate cancer
.
Clin Cancer Res
2005
;
11
:
3353
62
.
17.
Park
SS
,
Dong
H
,
Liu
X
,
Harrington
SM
,
Krco
CJ
,
Grams
MP
, et al
PD-1 restrains radiotherapy-induced abscopal effect
.
Cancer Immunol Res
2015
;
3
:
610
9
.
18.
Liu
X
,
Gibbons
RM
,
Harrington
SM
,
Krco
CJ
,
Markovic
SN
,
Kwon
ED
, et al
Endogenous tumor-reactive CD8+ T cells are differentiated effector cells expressing high levels of CD11a and PD-1 but are unable to control tumor growth
.
Oncoimmunology
2013
;
2
:
e23972
.
19.
Dronca
RS
,
Liu
X
,
Harrington
SM
,
Chen
L
,
Cao
S
,
Kottschade
LA
, et al
T cell Bim levels reflect responses to anti–PD-1 cancer therapy
.
JCI Insight
2016
;
1
:
e86014
.
20.
Yan
Y
,
Cao
S
,
Liu
X
,
Harrington
SM
,
Bindeman
WE
,
Adjei
AA
, et al
CX3CR1 identifies PD-1 therapy-responsive CD8+ T cells that withstand chemotherapy during cancer chemoimmunotherapy
.
JCI insight
2018
;
3
:
e97828
.
21.
Garg
I
,
Nathan
M
,
Packard
A
,
Kwon
ED
,
Larson
NB
,
Lowe
V
, et al
11 C-choline positron emission tomography/computed tomography for detection of disease relapse in patients with history of biochemically recurrent prostate cancer and prostate-specific antigen ≤0.1 ng/ml
.
J Cancer Res Ther
2021
;
17
:
358
65
.
22.
Mitchell
CR
,
Lowe
VJ
,
Rangel
LJ
,
Hung
JC
,
Kwon
ED
,
Karnes
RJ
. 
Operational characteristics of 11 C-choline positron emission tomography/computerized tomography for prostate cancer with biochemical recurrence after initial treatment
.
J Urol
2013
;
189
:
1308
13
.
23.
Scher
HI
,
Morris
MJ
,
Stadler
WM
,
Higano
C
,
Basch
E
,
Fizazi
K
, et al
Trial design and objectives for castration-resistant prostate cancer: updated recommendations from the prostate cancer clinical trials working group 3
.
J Clin Oncol
2016
;
34
:
1402
18
.
24.
Evans
JD
,
Morris
LK
,
Zhang
H
,
Cao
S
,
Liu
X
,
Mara
KC
, et al
Prospective immunophenotyping of CD8+ T cells and associated clinical outcomes of patients with oligometastatic prostate cancer treated with metastasis-directed SBRT
.
Int J Radiat Oncol Biol Phys
2019
;
103
:
229
40
.
25.
Hussain
M
,
Goldman
B
,
Tangen
C
,
Higano
CS
,
Petrylak
DP
,
Wilding
G
, et al
Prostate-specific antigen progression predicts overall survival in patients with metastatic prostate cancer: data from Southwest Oncology Group Trials 9346 (Intergroup Study 0162) and 9916
.
J Clin Oncol
2009
;
27
:
2450
6
.
26.
Contal
C
,
O'Quigley
J
. 
An application of changepoint methods in studying the effect of age on survival in breast cancer
.
Comput Stat Data Anal
1999
;
30
:
253
70
.
27.
Benedict
SH
,
Yenice
KM
,
Followill
D
,
Galvin
JM
,
Hinson
W
,
Kavanagh
B
, et al
Stereotactic body radiation therapy: the report of AAPM Task Group 101
.
Med Phys
2010
;
37
:
4078
101
.
28.
Ryu
S
,
Pugh
SL
,
Gerszten
PC
,
Yin
F-F
,
Timmerman
RD
,
Hitchcock
YJ
, et al
RTOG 0631 phase 2/3 study of image guided stereotactic radiosurgery for localized (1–3) spine metastases: Phase 2 results
.
Pract Radiat Oncol
2014
;
4
:
76
81
.
29.
Gomez
DR
,
Blumenschein
GR
,
Lee
JJ
,
Hernandez
M
,
Ye
R
,
Camidge
DR
, et al
Local consolidative therapy versus maintenance therapy or observation for patients with oligometastatic non-small-cell lung cancer without progression after first-line systemic therapy: a multicentre, randomised, controlled, phase 2 study
.
Lancet Oncol
2016
;
17
:
1672
82
.
30.
Palma
DA
,
Olson
R
,
Harrow
S
,
Gaede
S
,
Louie
AV
,
Haasbeek
C
, et al
Stereotactic ablative radiotherapy for the comprehensive treatment of oligometastatic cancers: long-term results of the SABR-COMET phase II randomized trial
.
J Clin Oncol
2020
;
38
:
2830
8
.
31.
Ost
P
,
Reynders
D
,
Decaestecker
K
,
Fonteyne
V
,
Lumen
N
,
De Bruycker
A
, et al
Surveillance or metastasis-directed therapy for oligometastatic prostate cancer recurrence (STOMP): five-year results of a randomized phase II trial.1. Ost, P. Surveillance or metastasis-directed therapy for oligometastatic prostate cancer recurren
.
J Clin Oncol
2020
;
38
:
10
.
32.
Chmura
S
,
Winter
KA
,
Robinson
C
,
Pisansky
TM
,
Borges
V
,
Al-Hallaq
H
, et al
Evaluation of safety of stereotactic body radiotherapy for the treatment of patients with multiple metastases
.
JAMA Oncol
2021
;
7
:
845
52
.
33.
Connor
MJ
,
Winkler
M
,
Ahmed
HU
. 
Survival in oligometastatic prostate cancer—a new dawn or the will rogers phenomenon?
JAMA Oncol
2020
;
6
:
185
.
34.
Gao
RW
,
Olivier
KR
,
Park
SS
,
Davis
BJ
,
Pisansky
TM
,
Choo
R
, et al
Single-fraction stereotactic body radiation therapy versus conventionally fractionated radiation therapy for the treatment of prostate cancer bone metastases
.
Adv Radiat Oncol
2019
;
4
:
314
22
.
35.
Wu
D
,
Yan
Y
,
Wei
T
,
Ye
Z
,
Xiao
Y
,
Pan
Y
, et al
An acetyl-histone vulnerability in PI3K/AKT inhibition-resistant cancers is targetable by both BET and HDAC inhibitors
.
Cell Rep
2021
;
34
:
108744
.
36.
Miyahira
AK
,
Den
RB
,
Carlo
MI
,
de Leeuw
R
,
Hope
TA
,
Karzai
F
, et al
Tumor cell heterogeneity and resistance; report from the 2018 Coffey-Holden Prostate Cancer Academy Meeting
.
Prostate
2019
;
79
:
244
58
.
37.
Dronca
RS
,
Mansfield
AS
,
Liu
X
,
Harrington
S
,
Enninga
EA
,
Kottschade
LA
, et al
Bim and soluble PD-L1 (sPD-L1) as predictive biomarkers of response to anti-PD-1 therapy in patients with melanoma and lung carcinoma
.
J Clin Oncol
2017
;
35
:
11534
.