The Potential Contribution of Radiopharmaceutical Therapies in Managing Oligometastatic Disease

OVERVIEW OF METASTASIS-DIRECTED THERAPY (MDT) IN PROSTATE CANCER

To this end, MDT has emerged as an attractive option for the growing population of patients diagnosed with molecularly defined mHSPC. The premise for why MDT might significantly impact natural history, rather than simply provide local control for treated lesions, stems from the discovery that metastasis-to-metastasis spread in the same patient is common, either through de novo monoclonal seeding of daughter metastases or through the transfer of multiple tumor clones between metastatic sites (15). Indeed, in contrast to the traditional belief in solid oncology that the cancer is no longer curable once it becomes metastatic, aggressive MDT to eliminate all sites of macroscopic disease in a patient with oligometastatic disease has been shown to lead to long-term disease control and possibly even a cure in certain cases (16–25).

The development of stereotactic body radiation therapy (SBRT), which involves the delivery of an ablative dose of radiation precisely to the lesions in 5 or fewer treatment sessions, is a critical tool for MDT. Given its biologic advantage of a higher dose per fraction, increased convenience with a shorter treatment course, a high local control rate, and a modest toxicity profile, SBRT has become the radiation modality of choice when delivering MDT. The randomized SABR-COMET trial is one of the earlier trials that tested whether MDT delivered via SBRT could improve outcomes in patients with oligometastatic disease. In this trial, 99 patients with controlled primary malignancies of various histologies who had 5 or fewer metastatic lesions were randomized 2:1 to SBRT to all sites of disease or the palliative standard of care (which included non-SBRT palliative radiation). The most common primary tumor types were breast (n = 18), lung (n = 18), colorectal (n = 18), and prostate (n = 16). At a median follow-up of 25 mo, median OS was 41 mo among patients receiving SBRT versus 28 mo in patients receiving the standard of care (P = 0.09) (26). With longer-term follow-up (median, 51 mo), SBRT was still associated with an increased OS (5-y OS, 42.3% vs. 17.7%; P = 0.006) and progression-free survival (PFS) (5-y PFS, not reached vs. 17.3%; P = 0.001) (17). However, an increased risk of grade 2 toxicity or higher was seen (29% vs. 9%, P = 0.026).

For prostate cancer specifically, MDT for oligorecurrent disease has been evaluated in 4 prospective studies, including 2 randomized phase II trials (Table 1). The STOMP trial enrolled 62 men with oligorecurrent disease after prior surgery or radiation who had no more than 3 metastases visible on ¹¹C-choline PET and randomized them to observation versus MDT (18). The primary endpoint was ADT-free survival, with ADT starting at the time of polymetastatic progression, local progression, or symptoms. PFS was a composite secondary endpoint, defined by biochemical progression (as per PCWG2 (27)), RECIST-based local progression (28), distant progression, and death from any cause. Among patients who received MDT, 81% received SBRT. Initial results at a median follow-up of 36 mo showed that MDT improved ADT-free survival from 13 to 21 mo and improved the median time to biochemical progression from 6 to 10 mo. In an update with longer follow-up, the PFS benefit of MDT was maintained (hazard ratio [HR], 0.48; P = 0.01) (29).

The ORIOLE trial enrolled 54 patients with oligorecurrent disease after prior surgery or radiation who had no more than 3 metastases visible on conventional imaging (19). The primary endpoint was the proportion of men with disease progression at 6 mo, defined as a composite endpoint of a PSA rise of at least 2 ng/dL and 25% above nadir; concern for radiologic progression by either CT, MRI, or bone scanning; symptomatic progression of disease; initiation of ADT for any reason; or death. All patients underwent ¹⁸F-DCFPyL PSMA PET/CT as well, though patients and investigators were masked to the results. Overall, SBRT reduced the proportion of men with disease progression from 61% to 19% at 5 mo (P = 0.005). With a median follow-up of 18.8 mo, the median PFS with SBRT was not reached, whereas it was 5.8 mo with observation (HR, 0.3; P = 0.002). Among the 36 men who underwent PSMA PET/CT, 16 (44.4%) had lesions not seen on conventional imaging. The proportion of progression at 6 mo was only 5% for men with no untreated PSMA-positive lesions, versus 38% for those with any untreated PSMA-positive lesions (P = 0.03). Median distant metastasis-free survival was 29.0 mo in men with no untreated lesions, versus 6.0 mo in men with any untreated lesion (P < 0.001).

In an update with a median follow-up of 5.3 y, the PFS benefit of MDT was maintained (HR, 0.48; P = 0.01) (29). In a per-protocol analysis, MDT improved castration-resistant prostate cancer–free survival (HR, 0.51; P = 0.12) (30).

A pooled analysis of both trials with a median follow-up of 52.5 mo found that MDT significantly improved PFS from 5.9 to 11.9 mo, with a pooled HR of 0.44 (P < 0.001) (29). However, radiographic PFS, time to castration-resistant prostate cancer, and OS were not improved. Patients whose tumors harbored a high-risk mutational signature (defined by pathogenic somatic mutations within ATM, BRCA1/2, Rb1, and TP53) had a shorter PFS, though these patients had a significantly larger PFS benefit from MDT as well.

Two additional single-arm phase II trials have used advanced molecular imaging to investigate MDT in patients with oligorecurrent disease. The POPSTAR trial enrolled 33 men who had up to 3 bony or lymph node lesions seen on a screening ¹⁸F-NaF-PET/CT scan (20). The primary endpoints were feasibility and tolerability; secondary outcomes included local and distant PFS, with the former scored by RECIST (28). Local PFS was 93% at 2 y, whereas 2-y distant metastasis-free survival was 39%. Twenty-two of the patients had hormone-sensitive disease, and among these patients, the freedom from ADT was 48% at 24 mo. The PSMA MRgRT trial enrolled patients with negative results on conventional imaging and 5 or fewer lesions on ¹⁸F-DCFPyL PSMA PET/MRI/CT who had a rising PSA of 0.4–3.0 ng/mL (23). Ultimately, 37 patients received MDT. At a median follow-up of 15.9 mo, 22% of patients had a complete biochemical response (PSA reduced to <0.05 ng/mL) and 60% of patients had a PSA decline of at least 50%. The median time to PSA progression was 17.7 mo. An update of the initial PSMA MRgRT cohort found that, with an extended follow-up to 40.7 mo, the biochemical response rate was maintained at 59.4%, with a complete biochemical response in 27% (31). In a validation cohort of 37 patients with identical enrollment criteria and a median follow-up of 14.3 mo, a biochemical response was seen in 43% of patients, with a complete response in 13.5%.

Taken together, studies confirm that MDT (predominantly delivered with SBRT) for patients with oligorecurrent prostate cancer significantly delays PSA-based progression. Importantly, MDT is safe: across these trials, of the 165 patients who received MDT, only 2 grade 3 toxicities attributable to MDT were seen. This contrasts with SABR-COMET, in which 3 treatment-related deaths were seen. The difference may be explainable by the fact that patients on the 4 prostate trials rarely had visceral metastases, which can be more challenging to irradiate.

Despite the favorable safety profile and overall initial efficacy, however, there are clearly patients who progress after MDT and may benefit from further treatment intensification (32,33). In a retrospective study of 258 patients with oligorecurrent mHSPC who had a median follow-up time of 25.2 mo, the median time to PSA recurrence after MDT was 15.7 mo and the median distant metastasis-free survival was 19.1 mo. Among patients who did not receive ADT, the median time to PSA recurrence was 10.9 mo and distant metastasis-free survival was 12.4 mo. Another 20 men were treated with a defined course of ADT; after stopping ADT, the median biochemical PFS was 17.6 mo. Overall, bone-only recurrence was the most common form of failure (44.2% of patients with recurrences), with another 24.8% of recurrences involving osseous disease in addition to another site. Node-only recurrences accounted for 26.5% of recurrences. Interestingly, the original site of recurrence was associated with subsequent sites of recurrence. Among patients treated for a bone lesion, most recurrences (86.5%) involved at least 1 osseous structure. For patients treated for a node-only lesion, most recurrences were also node-only (64.5%), though an osseous component was seen in 32.3% of recurrences. Three modes of progression were defined. Class I progressors, accounting for 40.9% of patients overall and 27.6% treated without ADT, had long-term control with no recurrences after 18 mo. Class II progressors, or oligoprogressors, had no more than 3 lesions at recurrence and accounted for 36% of patients overall and 44.8% of those treated without ADT. Class III progressors, or polyprogressors, had more than 3 lesions at recurrence and accounted for 23.1% of patients overall and 27.6% of those treated without ADT. Among patients who had advanced molecular imaging for follow-up, rather than conventional imaging, a lower percentage had long-term control (36.3%) and a higher percentage had polyprogression (26%). Overall, these data suggest that systemic therapy intensification is warranted in some patients with oligorecurrent mHSPC. Given that most men with early oligometastatic disease defined by molecular imaging may be seeking to avoid ADT—which may be primarily cytostatic in this context—alternative forms of systemic intensification warrant investigation.

OVERVIEW OF RADIOLIGAND THERAPY (RLT) AND THERANOSTIC THERAPIES IN PROSTATE CANCER

RLT, also known as radionuclide therapy, refers to the systemic administration of radiolabeled drugs targeting proteins that are specific to abnormal cells, allowing the delivery of localized radiation at the cellular level (34). Radioligands can be broadly classified into α-emitting radioligands and β-emitting radioligands. α-particles have short pathlengths of 50–80 μm, possess a linear energy transfer of 100 keV/μm, and can cause significant direct DNA damage (35,36). β-particles have a longer pathlength of 0.05–12 mm as well as lower linear energy transfer of 0.2 keV/μm (37). β-particles may thus be less directly efficacious, particularly against smaller lesions, and may have more toxicity against nontarget tissues (38). Overall, only 1 α-emitter is approved for clinical use, whereas multiple β-emitters are approved across different cancers. Specifically in the context of prostate cancer, 2 older β-emitters were approved for palliative use, but 1 α-emitter, ²²³Ra-dichloride (²²³Ra), and 1 β-emitter, ¹⁷⁷Lu vipivotide tetraxetan (¹⁷⁷Lu-PSMA-617), have been shown to improve OS (39). Studies with these agents are summarized below and in Table 2.

²²³Ra is a targeted α-emitter that, as a calcium mimetic, is preferentially incorporated into the bony matrix in areas of high bone turnover such as osteoblastic or sclerotic metastases (40–42).²²³Ra is thus an attractive RLT for metastatic castration-resistant prostate cancer (mCRPC), a lethal, end-stage form of prostate cancer in which bone-related complications are a leading cause of death (43). The benefit of ²²³Ra was shown in the ALSYMPCA trial (44). In this trial, 921 men with progressive mCRPC and 2 or more symptomatic bone metastases with no known visceral metastases were randomized in a 2:1 fashion to receive either 6 doses of ²²³Ra (dosed at 50 kBq/kg of body weight every 4 wk) or placebo, in addition to the standard of care. The primary endpoint was OS, and this was improved by the addition of ²²³Ra (median OS, 14.9 vs 11.3 mo; P < 0.001). The time to the first symptomatic skeletal event was significantly prolonged as well (median, 15.6 vs. 9.8 mo; P < 0.001). No significant differences in adverse events of grade 3 or higher were noted between arms. At the median follow-up of 13 mo, no acute myeloid leukemia, myelodysplastic syndrome, or new primary bone cancers were seen (45). Quality of life was also assessed; using an estimation of pain-related symptoms based on the Functional Assessment of Cancer Therapy–Prostate questionnaire, patients receiving ²²³Ra were found to be more likely to experience meaningful improvements in pain (30.2% vs. 20.1%; P = 0.010) (46). An earlier randomized phase II trial using a similar dosing regimen, but with only 4 doses total, identified a benefit in terms of time to first bone-related event as well (47). A trial-level metaanalysis that included both studies found a pooled HR of 0.70 (95% CI, 0.58–0.83) for improving OS (48). Unfortunately, the addition of ²²³Ra to abiraterone acetate with prednisone failed to improve symptomatic skeletal event–free survival or OS in the evaluation of ²²³Ra in combination with abiraterone in castration-resistant prostate cancer (ERA 223) trial (49). Significantly more fractures were seen in patients receiving ²²³Ra (9% vs. 3%), particularly in patients not receiving bone protection agents.

Theranostics is a precision medicine approach that uses targeted radioactive compounds to image specific cell surface markers and subsequently uses RLTs to irradiate tissues expressing these markers (50). ¹⁷⁷Lu-PSMA-617 is a chemically modified DOTA-conjugated PSMA binder that has allowed the first theranostic therapy in prostate cancer. In the VISION trial, 831 patients with mCRPC and at least 1 PSMA-positive lesion were randomized in a 2:1 ratio to either 4 or 6 cycles of 7.4 GBq of ¹⁷⁷Lu-PSMA-617 every 6 weeks with standard of care versus standard of care alone (51). Specific eligibility criteria were at least 1 PSMA-positive lesion with uptake greater than liver parenchyma and no large PSMA-negative lesions, disease progression after treatment with at least 1 second-generation androgen receptor signaling inhibitor and 1 or 2 taxanes, and life expectance of more than 6 mo. Treatment with ¹⁷⁷Lu-PSMA-617 improved OS (median, 15.3 vs. 11.3 mo; P < 0.0001) and radiographic PFS (median, 8.7 vs. 3.4 mo; P < 0.001). Grade 3 or higher adverse events were higher in the experimental group (52.7% vs. 38.0%), but overall quality of life was not impacted. The most common adverse events included fatigue, dry mouth, anemia, and back pain. Time to first symptomatic skeletal event or death was also prolonged (median, 11.5 vs. 6.8 mo; P < 0.001). A subsequent quality-of-life analysis found that time to worsening of quality of life by the Functional Assessment of Cancer Therapy–Prostate metric was prolonged with ¹⁷⁷Lu-PSMA-617 (52). Hematologic adverse events of grade 3 or higher included decreased hemoglobin (15% vs. 6%), lymphocyte concentrations (51% vs. 19%), and platelet counts (9% vs 2%).

The phase II TheraP trial randomized 200 patients with mCRPC and prior docetaxel treatment to ¹⁷⁷Lu-PSMA-617 or cabazitaxel (53). Patients were required to have at least 1 ⁶⁸Ga-PSMA–positive lesion with an SUV_max of at least 20 (with all other PSMA-avid sites having an SUV_max of ≥10) and nondiscordant findings between PSMA and ¹⁸F-FDG PET/CT. The dosage schedule for ¹⁷⁷Lu-PSMA-617 was 8.5 GBq for the first cycle with a 0.5-GBq decrease per each subsequent cycle (maximum of 6 cycles with 6 wk between cycles). The primary endpoint was the PSA response rate (≥50% reduction), and treatment with ¹⁷⁷Lu-PSMA-617 did significantly improve this (66% vs. 37%, P < 0.0001). The effect of treatment on PFS was not constant with time, and the impact appeared to be more pronounced after 6 mo. However, when a Cox model was used, progression was delayed with ¹⁷⁷Lu-PSMA-617 (HR, 0.63; P = 0.0028). ¹⁷⁷Lu-PSMA-617 did not increase the rate of overall toxicities of grade 3 or higher (33% vs. 53%), though thrombocytopenia was more common with ¹⁷⁷Lu-PSMA-617 (11% vs. 0%). Improvements in quality of life and symptoms were seen with ¹⁷⁷Lu-PSMA-617 with respect to diarrhea, fatigue, social functioning, and insomnia, and deterioration-free survival for global health status was better for men receiving ¹⁷⁷Lu-PSMA-617 at 6 mo (9% vs. 13%, P = 0.0002). A post hoc analysis found that ⁶⁸Ga-PSMA-PET SUV_mean was predictive of a higher likelihood of a favorable response and a high ¹⁸F-FDG PET metabolic tumor volume associated with a lower response regardless of randomly assigned treatment (54).

¹⁷⁷Lu-PSMA-I&T (also known as ¹⁷⁷Lu-PNT2002) is the second PSMA-targeting RLT that has been studied in large clinical trials. It has more kidney uptake, but less lacrimal uptake, than ¹⁷⁷Lu-PSMA-617 (55–57) and has shown anticancer activity in the compassionate-use setting for patients with heavily pretreated mCRPC (58). It is being studied in 2 phase III randomized trials in the mCRPC space: SPLASH (NCT04647526) and ECLIPSE (NCT05204927). Preliminary results from both were expected in late 2023.

COMBINING RLT WITH MDT

Overall, the data suggest that although MDT for mHSPC is effective at controlling individual lesions, its potential as a curative option is limited because of the existence of occult disease at the time of treatment. The use of advanced molecular imaging for patient selection may increase the percentage of patients with a long-term response, but ultimately most patients will still experience progression. RLT possesses significant antineoplastic activity even in the most advanced setting of mCRPC but ultimately is not a curative option given the natural history of CRPC and a limit to the number of cycles that can safely be administered (59). The ultimate limitation after RLT may also depend on the emergence of PSMA-negative, ¹⁸F-FDG–positive disease that is no longer effectively targeted. Earlier administration of RLT therapy before such clones can emerge and when the burden of disease is lower may increase the effectiveness of RLT in durably controlling disease. A logical synergy between MDT and the theranostic approach might be achieved using both MDT and RLT in patients with oligorecurrent mHSPC. The disease setting would by definition include a low burden of disease, and we would not expect the presence of PSMA-negative ¹⁸F-FDG–positive occult disease, which would improve the efficacy of the RLT. Similarly, the addition of RLT and the use of imaging to select patients with lower-volume disease would improve the efficacy of MDT by reducing the rates of oligo- and polyprogression.

The potential benefit of this synergy also rests on the hypothesis that RLT would be effective, without combination with ADT, in mHSPC. This was tested in the randomized multicenter phase II BULLSEYE trial (60). In this trial, men with mHSPC and 5 or fewer lesions, with an SUV_max of more than 15 for all lesions and a PSA doubling time of no more than 6 mo were randomized in a 1:1 fashion to 2–4 cycles of 7.4 GBq of ¹⁷⁷Lu-PSMA-617 versus deferred ADT. The primary endpoint is progression within 24 wk of cycle 2, with progression defined as a 100% increase in PSA or radiographic or clinical progression. Early results based on the first 42 patients enrolled on the study indicated a promising effect from the treatment (34). The median PSA was 4.5 ng/dL at inclusion. At 6 mo of follow-up, 10% of patients on the treatment arm versus 77% of patients on the control arm had experienced progression. The median PFS was not reached in the treatment arm, versus 4 mo in the control arm (P < 0.001). Overall, 24% of patients had a complete biochemical and imaging response. Only 3 grade 3 adverse events were seen. Though early, these interim results support the concept that RLT agents are active in the setting of mHSPC as well.

The direct question of whether the integration of RLTs with MDT will improve outcomes in mHSPC is being tested in 3 randomized phase II trials and 2 single-arm phase 2 studies. These are summarized in Table 3. The RAVENS and LUNAR trials will be fully accrued by the end of 2023. The phase III PSMA-DC study (NCT05939414) is also planned to open to accrual in late 2023 for patients with only molecular oligometastatic disease. As data supporting RLT in earlier disease states matures, more studies integrating RLT with MDT are likely on the horizon.

DISCLOSURE

Amar Kishan received personal fees from ViewRay, Inc., Varian Medical Systems, Inc., and Janssen Pharmaceuticals outside the submitted work, as well as research funding from ViewRay, Janssen Pharmaceuticals, PointBiopharma, and Lantheus. He reports research funding from grant PC210066 from the Department of Defense. Michael Hofman reports financial support from Novartis (including AAA and Endocyte), AstraZeneca, Bayer Corp., Isotopia, and MIM Software Inc.; a consulting or advisory relationship with Astellas Pharma Inc.; and consulting fees for lectures or advisory boards from Astellas and AstraZeneca in the last 2 y and from Janssen, Merck/MSD, and Mundipharma in the last five years. Ana Kiess has institutional clinical research funding from Bayer, Novartis/AAA, and Merck and has participated in an unpaid consulting and advisory board for Novartis. Phuoc Tran received a research grant from Bayer; consulted for Bayer, RefleXion, Myovant, Natsar Pharmaceuticals, Regeneron, Lantheus, and Janssen; and has a patent licensed to Natsar Pharmaceuticals. Jeremie Calais reports prior consulting services for Advanced Accelerator Applications, Astellas, Blue Earth Diagnostics, Curium Pharma, DS Pharma, EXINI, GE Healthcare, Isoray, IBA RadioPharma, Janssen Pharmaceuticals, Lightpoint Medical, Lantheus, Monrol, Novartis, Progenics, POINT Biopharma, Radiomedix, Sanofi, and Telix Pharmaceuticals outside the submitted work. No other potential conflict of interest relevant to this article was reported.