Systemic therapy escalation after stereotactic body radiation therapy for oligometastatic hormone-sensitive prostate cancer

Graphical abstract

Highlights

• •Androgen deprivation therapy is the standard treatment for prostate cancer metastatic disease.
• •Androgen deprivation therapy is associated with numerous side effects.
• •For oligometastatic disease, stereotactic radiation therapy could postpone systemic treatment.
• •This large cohort results strengthen that SBRT allows to postpone systemic therapy for omHSPC patients.

Abstract

Purpose

To evaluate the oncological outcome after stereotactic body radiation therapy (SBRT) for oligometastatic hormone-sensitive prostate cancer (omHSPC) patients.

Materials-Methods

In this retrospective, observational, multi-institutional study, omHSPC patients (≤5 metastases) underwent SBRT. Primary endpoint was systemic therapy escalation-free survival (STE-FS) after SBRT. Local (LR), distant (DR), prostatic (PR) and isolated biochemical (iBR) relapses were reported with progression-free survival (PFS) and overall survival (OS). Prognostic factors for STE-FS were investigated. Toxicity was reported.

Results

From 01/07 to 09/19, 119 pts with omHSPC underwent SBRT. With a MFU of 34 months [12–97], median STE-FS was 33.4 months (95%CI 26.6–––40.1). Median OS was not reached and PFS was 22.7 months (CI95% 18.6–––32.3). Post-SBRT-PSA remained stable or decreased in 87 pts (73.1%). Progression events (LR, MR, PR, iBR) were observed in 72 pts (60.5%), among whom 6 relapsed in the irradiated area (local control rate: 95%). DR, BR, PR were observed in 44 pts (37%), 21pts (17.7%) and 2 pts (1.7%) respectively. In multivariate analysis, post-SBRT biochemical response was an independent prognostic factor for STE-FS. Grade ≥ 3 toxicity occurred in 1pt.

Conclusion

With excellent local control and tolerance, SBRT for omHSPC patients represents an attractive approach to defer systemic therapeutic escalation and prevent its side effects. Accurate patient selection for SBRT requires more data with longer follow-up and higher numbers of patients pending the results of upcoming randomized trials.

1. Introduction

In 2020, prostate cancer remained the leading cause of cancer in men worldwide, with 1.41 million cases and a mortality of more than 375,000 patients [1]. Mainly due to more accurate detection techniques, the rate of patients with metastatic disease, whether synchronous or metachronous, continues to increase with an expected annual burden of + 42% in 2025 in the United States [2]. Faced with this public health issue, improvement of therapeutic management of patients with metastatic prostate cancer appears crucial.

Since the discovery of androgen ablation beneficial effects [3], androgen deprivation therapy (ADT) based on Luteinizing hormone-releasing hormone analogs has been the treatment of choice, even in low lesion burden situations [4]. Recently, high proof levels led to consideration of a-LHRH based ADT plus Docetaxel and/or novel hormonal therapies (NHT) as a new standard of care for metastatic prostate cancer [5], [6], [7]. In addition, the combination of ADT with prostate irradiation significantly improved oncological outcome especially for patients with low burden metastatic disease [8].

Despite these improvements, metastatic prostate cancer remains a life-threatening condition, with overall survival (OS) rates of 5.3 years and 4.3 years for low and high-burden metastatic disease respectively, for patients managed with ADT plus docetaxel [9]. Moreover, the effectiveness of systemic therapies remains limited in time. A hormone-sensitive metastatic prostate cancer (mHSPC) patient undergoing ADT has a median time of about 30 months before becoming castration-resistant [10], while the median OS for castration-resistant metastatic prostate cancer (mCRPC) is about 27 months [11]. Faced with the limited number of systemic medications available, their respective side effects and their reduced duration of efficacy, systemic therapy escalation (STE) is a significant milestone in prostate cancer progression.

Reducing the duration of exposure to ADT in mHSPC did not show a significant reduction in time to disease progression [12], [13]. However, these studies did not consider the number of metastases, and more specifically, patients with an oligometastatic hormone-sensitive prostate cancer (omHSPC). By locally controlling a small number of metastases, SBRT could temporarily avoid STE, and preserve the systemic treatments available while reducing the occurrence of side effects. This study analyzed the oncological outcome after SBRT for omHSPC patients.

2. Material and methods

In this retrospective, observational, multi-institutional study (Nice, Lille, Monaco), collected databases of patients with omHSPC who underwent SBRT were reviewed. Consent of each patient was obtained prior to analysis, after providing clear and fair information on the use of the data. In accordance with current legislation, data collection was registered at the National Health Data Hub under the number N° F20210402112942.

2.1. Patient features

Selection criteria were oligometastatic (≤5 metastases) hormone-sensitive prostate cancer patients [12] treated with ablative SBRT on at least one of the metastatic sites (bone, pelvic (PLN - below the promontory) or extra-pelvic lymph node (EPLN), or viscera). The imaging workup was left to the discretion of participating centers. After SBRT, patients were followed for at least 12 months, according to the usual modalities of each center (clinical examination, blood test and/or imaging). Systemic therapy by short-course ADT could be prescribed concomitantly with SBRT.

Initial prostate cancer characteristics, patient epidemiological data, prostate specific antigen (PSA) kinetics and specific therapies before first SBRT were reported. Patients could have a history of synchronous (metastatic disease discovered within 3 months after PC diagnosis) or metachronous metastatic disease. An outline of the patients' disease history is provided in Fig. 1.

Fig. 1.

Schematic timeline of patient’s oligometastatic disease history.

2.2. SBRT technique

SBRT had to be performed using a device allowing delivery of a high dose in a small volume, with a strong dose gradient, with a maximum of 10 fractions. Methods of delineation, delivery and performance of the irradiation were to be in accordance with good clinical practice. All SBRT regimens were considered. Data related to total dose, fractionation, total treatment time, and irradiation site were collected.

2.3. Patient follow-up

Primary endpoint was systemic therapy escalation-free survival (STE-FS), corresponding to the time interval between SBRT first course and the next introduction or modification of a systemic treatment. Systemic therapies could be ADT (a-LHRH) or NHT (Abiraterone acetate + Prednisolone, Enzalutamide, Apalutamide or Darolutamide), or chemotherapy (Docetaxel or Cabazitaxel). The initiation or modification of systemic therapy was left to the discretion of the physicians at each center, who followed the EAU guidelines available at the time of management.

Secondary endpoints were overall-survival (OS) defined by the time interval between SBRT and death (from any cause), progression free-survival (PFS) corresponding to the time interval between SBRT and any prostate cancer progression events. Oncological progression events after SBRT were considered: local relapse (LR; i.e. in the SBRT field), distant metastatic relapse (DR), prostatic relapse (PR, i.e. in the prostate gland) or isolated biochemical relapse (iBR). LR, PR and DR were defined as the appearance of new lesions or progression of known lesions (RECIST analysis) combined with a rising PSA (≥25% over the last 12 months). Patients presenting a rising PSA without imaging relapse in the 6 following months were considered with iBR. The first three PSA blood tests after irradiation were collected. Based on the comparison between PSA at the time of SBRT (PSA_SBRT) and first PSA after SBRT (PSA₊₁), three different biochemical situations were observed: PSA response (PSA_SBRT-PSA₊₁ greater than 0.2 +/- 0.1 ng/mL), PSA stable (PSA_SBRT-PSA₊₁ = 0 +/- 0.1 ng/mL) and PSA progression (PSA_SBRT-PSA₊₁ < -0.2 +/- 0.1 ng/mL).

Prognostic factors of STE-FS were analyzed. Toxicity (acute and late) was assessed (CTCAE v5.0)

2.4. Statistical analysis

Results for primary and secondary endpoints were presented with a 95% confidence interval (95%CI, Rothman for survival data). Qualitative data were presented as absolute frequency, relative frequency, 95% confidence interval, percentage of missing data. These data were compared using the Chi2 test or Fisher's test in case of non-compliance with the Chi2 conditions.

Quantitative data were presented as a histogram, median, extreme and standard deviation. The normality of these parameters was evaluated using frequency histograms and Shapiro's test. Simple mathematical transformations were used to normalize non-normal data. Quantitative data were compared using Student's T test or Mann-Withney test in case of non-compliance with the conditions of application of Student's test.

Censored data were defined between the date of inclusion and the date of occurrence of the event, patients lost to follow-up were censored at the date of last news. These data were presented graphically as Kaplan Meier curves, survival rates at various times, median survival and 95%CI of the study population were shown. The survival curves were compared by the Log-Rank test. All statistical analyses were performed using R.3.0.2 software for Windows.

3. Results

3.1. Patients and treatment features

From January 2007 to September 2019, 119 pts (145 metastatic lesions) underwent SBRT. At the time of primary prostate cancer diagnosis, 109 pts (91.6%) were localized while 10 pts (8.4%) also had synchronous metastatic disease. In the latter group, only one patient received initial management SBRT consisting in external body radiation therapy (EBRT) on the prostate plus concomitant and adjuvant ADT. Among the 109 pts with localized prostate cancer, 88 pts (80.7%) underwent a primary surgical procedure, associated with adjuvant or salvage radiation therapy for 18 pts and 11 pts respectively (e-Table1, Supplementary data).

Table 1.

Patient (n = 119) and treatment descriptions at the time of SBRT.

Items	Data	% / [min - max]
Median ageSBRT (year)	70.5	[53.3–––100.5]
Median time between initial diagnosis and metastatic disease	69	[1.9–––202]
Chronology of metastasis diagnosis (# pts)
Metachronous metastatic disease	109	91.6
Synchronous metastatic disease	10	8.4
Total number of irradiated metastases	145
Number of irradiated metastasis/number of metastasis (# pts)
1/1	95	80
2/2	19	16
3/3	2	1.6
4/4	1	0.8
1/2	1	0.8
1/5	1	0.8
Metastatic disease irradiated sites (# pts)
Bone	64	53.8
PLN	47	39.5
EPLN	8	6.7
Median biological results at the time of SBRT
PSA (ng/ml)	2.1	[0.01 – 277]
PSADT (months)	5.9	[0.9 – 45.2]
PSAVelocity (ng/ml/y)	1.8	[0.1 – 15.0]
ADT + SBRT combination Yes	38	31.9
No	81	68.1
SBRT regimens
27 Gy / 3f	38	31.9
35 Gy / 5f	29	24.4
36 Gy / 6f	22	18.5
Others	30	25.2

MFU (months): Median follow-up; PSA_DT: Prostate specific antigen doubling time; PSA_Velocity: Prostate specific antigen velocity; PLN: Pelvic lymph node metastasis; EPLN: Extra-pelvic lymph node metastasis; # pts: number of patient; # metastases: number of metastases; age_SBRT: patient age at the time of SBRT; ADT: first generation androgen deprivation therapy; SBRT: Stereotactic body radiation therapy.

At the time of first SBRT, median age was 70.5 years [53.3–––100.5] (Table 1). Bone metastases represented the majority of the treated lesions (53.8%). Median number of radiologically visible metastases was 1 [1], [2], [3], [4], [5] with a median number of irradiated metastases of 1 [1], [2], [3], [4]. Over 98% of patients were treated at all oligometastatic sites (Table 1). One patient received SBRT on a lymph node site, while the second bone site was operated on. Another patient underwent SBRT on a bone site, while the other 4 lymph node sites were treated with definitive ADT. Median time from first metastasis to SBRT was 2.9 months [0.1–––74.3]. In 38 patients (31.9%), SBRT was associated with concomitant ADT with a median ADT duration of 7 months [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17].

Regarding irradiation procedures, the Nice and Lille Centers used Cyberknife® linac (Accuray® Sunnyvale, California, US) and the Monaco institute used Novalis® TrueBeam STx® linac (Varian® Palo Alto, California, US). Different irradiation regimens were used (total dose and fractionation). One hundred and one patients (98.3%) were treated with a 3 to 6 fraction protocol for a total dose ranging between 27 and 36 Gy (Table 1; e-Table 2, Supplementary Data). In the context of already irradiated territories, two patients had 10 fraction schedules. Median overall treatment time was 6 days [3–26].

3.2. Oncological outcome

With a median follow-up (MFU) of 34 months [12–97], 47 pts (39.5%) were free of prostate cancer progression. Regarding the primary endpoint, STE occurred in 57 patients (47.9%) leading to a median STE-FS of 33.4 months (95%CI [26.6 – 40.]) (Fig. 2). The introduction (first or reintroduction) of ADT was the most frequent STE event (Table 2). Only two patients underwent NHT introduction while they had been prescribed SBRT combined with short-course ADT, because of early progression during the first 6 months after SBRT.

Fig. 2.

Kaplan-Meier survival curves after SBRT on metastasis for oligometastatic prostate cancer patients (119 pts): OS (overall survival: red line); STE-FS (systemic therapy escalation-free survival: black line) and PFS (progression free survival: blue line). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 2.

Oncological outcome analysis after SBRT.

Items	Data	% / [min - max]
MFU (months)	34	[12–97]
Post-SBRT oncological outcome (# pts)
Non-progressive disease	47	39.5
Progressive disease	72	60.5
LR alone	4
LR + DR	1
LR + PR	1
DR	44
DR + PR	1
iBR	21
Post-SBRT biochemical response (# pts)
Response	77	64.7
Stable	10	8.4
Progression	20	16.8
Data missing	12	10.1
Post-SBRT STE (# pts)
No STE	62	52.1
STE	57*	47.9

# pts: number of patients; MFU: Median follow-up; SBRT: Stereotactic body radiation therapy; PSA₊₁: First prostate specific antigen after SBRT; PSA_SBRT: Prostate specific antigen at the time of SBRT; ADT: first generation androgen deprivation therapy; NHT: novel hormonal therapy; STE: systemic therapy escalation; LR: local relapse in the irradiated field; DR: distant metastatic relapse; PR: prostatic relapse; iBR: isolated Biochemical relapse.

Post-SBRT biochemical response: Response represents PSA₊₁ < PSA_SBRT; Stable represents PSA₊₁ = PSA_SBRT +/- 0.1 ng/mL; Progression represents PSA₊₁ > PSA_SBRT.

*54 patients started ADT and 2 NHT.

Seventy-two patients (60.5%) experienced progression with a median time to progression of 17.4 months [2.8 – 77.1]. Progressive disease was observed in 16 pts (42%) and 56 pts (69%) with and without SBRT concomitant ADT. Local control rate was 95% with 6 pts who underwent a local relapse in the SBRT field, of which 1pt had a synchronous prostatic relapse and 1pt presented a synchronous distant relapse. Irradiated sites were bone (5 pts) and PLN (1pt). For these 6 pts, the delivered doses were 35 Gy/5f (3 pts), 25 Gy/5f (1pt), 36 Gy/6f (1pt) and 30 Gy/5f (1pt). Distant metastatic relapse occurred in 45 pts, of which one had a synchronous prostatic relapse. Twenty-one isolated biochemical relapses have been registered (Table 2). Median time to first post-irradiation PSA (PSA₊₁) was 2 months [0.1 – 7]. PSA progression, stable and response were observed for 20 pts (16.8%), 10 pts (8.4%) and 77 pts (64.7%) respectively. Data were missing for 12 pts (10.1%). Median OS was not reached and PFS was 22.7 months (CI95% [18.6 – 32.3]) (Fig. 2). Among the 9 deaths (7.6%), 4 were related to prostate cancer.

In univariate and multivariate analysis, only the post-SBRT biochemical response was a prognostic factor for STE-FS (19.7 vs. 33.6 months for PSA progression vs. PSA stable/response; HR 0.46, 95%CI [0.26 – 0.87]; p < 0.017). STE-FS was not correlated with ADT + SBRT combination, metastatic status, number of metastases, PSA_Velocity or PSA_DT (e-Table 3).

Oncological outcome was investigated in a subgroup analysis according to the use of ADT + SBRT combination (38 pts) or not (81 pts). ADT + SBRT combination was associated with a significantly longer PFS (18.6 vs. 32.9 months; HR 0.5, 95%CI [0.29–––0.89]; p < 0.018), with no impact on STE-FS (29.7 vs.33.4 months; HR 0.7, 95%CI [0.38 – 1.3]; p = 0.247) and OS (medians not reached; HR 1.6, 95%CI [0.37 – 6.7]; p = 0.53) (Fig. 3).

Fig. 3.

Kaplan-Meier STE-free survival (A), PFS (B) and OS (C) curves after SBRT on metastasis for oligometastatic hormone-sensitive prostate cancer patients combined with ADT (38 pts, blue line) or not (81 pts, red line) at the time of SBRT. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.3. Toxicity profile

Only 1pt experienced grade ≥ 3 acute toxicity, while no late toxicity was observed. Radio-induced fracture of a vertebra occurred after stereotactic irradiation of a lumbo-aortic lymph node stuck to the spine (33 Gy in 3 fractions).

4. Discussion

Although systemic therapy represents the standard strategy for metastatic prostate cancer, its escalation for omHSPC reduces future therapeutic options, uses up the period of treatment effectiveness and exposes to risks of toxicities with deleterious impact on patient quality of life. To our knowledge, this is one of the largest retrospective reports on SBRT in omHSPC patients, confirming feasibility and safety of this treatment, allowing postponing systemic therapy. SBRT is presented as a therapeutic modality that makes it possible to defer this escalation by controlling metastatic disease progression, with an acceptable risk of toxicity.

In this study, median STE-FS was 33.4 months. For mHSPC patients, while a monitoring strategy (by waiting for the occurrence of significant oncological events) in terms of metastatic disease progression can delay ADT introduction by approximately 1 year [14], prospective studies demonstrated that SBRT can defer the introduction of ADT by approximately 2 years [14], [15], [16], [17] (e-Table 4, Supplementary data). On the other hand, for mHSPC patients, long-term LATITUDE trial results showed a median time to subsequent prostate cancer therapy of 33.7 months by using Abiraterone acetate plus Prendisolone while SBRT seems to offer comparable results to delay therapeutic escalation [6].

In the current study, PFS was 22.7 months, while the median OS was not reached. These results are in line with the ORIOLE phase II trial in which the authors reported a 12-month PFS of 70% after SBRT for mHSPC patients [18]. Furthermore, median PFS reported in our study appears comparable to that reported in the GETUG-AFU 15 trial with the addition of Docetaxel in mHSPC (22.9 months) [19]. Unfortunately, in our study, radiographic and biochemical PFS, frequently used in literature, could not be properly evaluated due to its retrospective nature.

In our cohort, with a MFU of 34 months, local control was 95%, comparable to the current results reported in literature [15], [20]. PSA kinetic (PSA_DT and PSA_Velocity) was not a prognostic factor for STE-FS. However, in the STOMP study, the authors reported that metastasis-directed therapies (including SBRT or surgery) had a significant impact on ADT-free survival for oligometastatic prostate cancer patients with PSA_DT < 3 months or lymph node metastatic locations [14]. Furthermore, Decaestecke et al., highlighted short PSA_DT as the only prognostic factor for ADT- free survival and PFS for oligo-progressive PC [16].

The potentiation of ionizing radiation by ADT for primary high-risk PC is well established; however, in the setting of mHSPC patients treated with SBRT, the benefit of this combination remains under debate, as well as its duration. In this context, a subgroup analysis was performed by comparing STE-FS, PFS and OS for patients with and without a combination of ADT + SBRT. Although PFS appeared significantly improved by an ADT/SBRT combination (18.6 vs. 32.9 months; p < 0.018), STE-FS and OS were non-significantly different (Fig. 3). This combination benefit should be qualified by the duration of ADT which could negatively impact patient quality of life. In the recent EXTEND phase II prospective trial [21], with a MFU of 22 months, the authors reported a PFS improvement in the ADT/SBRT arm (median not reached) compared to the ADT alone arm (15.8 months). These results suggest that ADT/SBRT combination could improve PFS compared to ADT alone while the use of SBRT as sole salvage therapy appears at least as good as ADT alone. Concerning the duration of hormone therapy after irradiation, the information from this multicenter study does not provide robust information. Moreover, time to castration-resistant prostate cancer status after SBRT was not collected, making it impossible to ascertain whether a concomitant ADT short-course affects the time of hormonal sensitivity.

Regarding SBRT safety for omHSPC, as it was already reported in literature, SBRT was associated with<3% of G ≥ 3 acute toxicity with no late observable side-effects [14], [15], [18] (e-Table 4, Supplementary data).

The main limitations of this study, in addition to its retrospective nature, are the heterogeneity of the different patient groups and their small number. Despite subgroup analyses, the use of ADT in combination with SBRT for over 30% of the patients limits interpretation. In addition, retrospective data collection did not allow us to determine the duration of this systemic treatment. Furthermore, the 12-year collection period represents another issue due to major changes in omHSPC patient management including imaging modalities, systemic treatments as well as the current use of SBRT in daily practice. Various imaging modalities have been used to define oligometastasis status: CT scan, bone scan, PSMA-PET, Choline-PET, Fluciclovine-PET. In our study, no clear data are available to assess the impact of staging imaging on the primary endpoint.

5. Conclusion

The management of omHSPC patients is undergoing an evolution rather than a revolution. In this setting, SBRT is becoming a major therapeutic tool by not only providing an excellent local control of metastatic disease but also by substantially delaying the introduction or modification of systemic therapy, without significant deleterious impact on OS. Furthermore, SBRT for omHSPC provides excellent safety and could have a positive impact on patient quality of life and possibly on health reimbursement systems. Patient selection for SBRT remains a crucial point to be defined. More consistent data with longer follow-up and higher numbers of patients are needed.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary material

The following are the Supplementary data to this article: