Sharing the burden: the case for definitive local therapy in place of immune checkpoint blockade for patients with a low-volume burden of metastatic disease

INTRODUCTION

The efficacy of immune checkpoint inhibitor (ICI) therapy depends on multiple factors including tumor PD-L1 expression, tumor mutational burden (TMB), and whether tumor DNA mismatch repair is intact or deficient. Evidence also suggests that overall tumor burden influences ICI efficacy, with lower tumor volume associated with improved outcomes.¹ Simultaneously, there is growing interest in managing limited, low-volume metastatic disease with definitive local therapy, defined here as ablation or resection of all sites of disease (both primary and metastatic) with the intent of providing maximal long-term local control. Prospective clinical trials in patients with oligometastatic cancer (usually defined as 1–6 sites of disease) have demonstrated a benefit to definitive local therapy either in terms of progression-free, overall, or systemic therapy-free survival.^2–12 Local treatments like surgery or SBRT demonstrate efficacy independent of factors such as PD-L1 status and TMB. These ablative treatments may also be less toxic than systemic therapies like ICI. Given that patients with a low overall tumor burden benefit more from ICI and given the success of definitive local therapy in patients with limited sites of disease, we suggest that there exists a sub-population of low-disease-burden patients who would experience outcomes with local therapy alone that are equivalent to those seen with ICI. Herein we review data supporting this notion, comparing the results of trials of definitive local therapy for oligometastatic cancer to those seen with current recommended first-line ICI-based therapies for metastatic cancer. We additionally recommend additional avenues for research to determine which patients will benefit from definitive local therapy.

TUMOR BURDEN and ICI EFFICACY

As reviewed by Dall’Olio and colleagues (2022), multiple analyses demonstrate that survival and response to ICI are independently associated with various measures of tumor burden.¹ These include RECIST measurement of tumors on CT imaging, metabolic tumor volume as seen on PET, and biomarkers such as circulating tumor DNA.¹ A secondary analysis of the KEYNOTE-001 trial of pembrolizumab for metastatic melanoma identified baseline tumor size (defined as the sum of the longest dimensions of all measurable baseline lesions) below 10.2 cm as associated with improved overall survival (OS).¹³ A retrospective cohort study investigating total metabolic tumor volume (tMTV, defined using the maximum standardized uptake value from 18-F-FDG PET scans) in non-small cell lung cancer (NSCLC) showed a significant association between increased tMTV and worse OS in patients receiving anti-PD-1 therapy but not in patients with EGFR driver mutations who received EGFR-targeted therapies.¹⁴ This result suggests that burden of disease could be predictive for ICI response as well as prognostic. These studies show that patients with few, discretely measurable lesions tend to respond best to ICI therapy. Data from randomized trials of definitive local therapy for oligometastases suggest that such patients may also benefit from definitive local therapy alone or in conjunction with systemic treatment.

METASTASIS-DIRECTED THERAPY FOR OLIGOMETASTASES in ICI-RESPONSIVE DISEASE

Immune checkpoint inhibitors are recommended as first-line therapy for NSCLC (without driver mutations), melanoma, and renal cell carcinoma (RCC).^15–17 There are also clinical trial data for the treatment of oligometastases arising from these malignancies, making for a useful comparison of local therapy to ICI, acknowledging the caveats inherent to comparing results across trials.

Several trials have investigated definitive local therapy for metastatic NSCLC with encouraging results. One paradigm uses induction chemotherapy to select candidates for local treatment. Gomez and colleagues investigated maintenance chemotherapy with or without surgery and/or SBRT for 1–3 metastases (plus the primary tumor, if previously untreated) in patients who did not progress after 3 cycles of induction chemotherapy. This trial identified a significant improvement in both median progression-free survival (PFS, 14.2 versus 4.0 months) and median OS (37.7 versus 17.0 months), with local treatment versus maintenance chemotherapy alone, respectively.^3,18 Another trial randomized NSCLC patients with 1–6 stable or partially responding sites of disease (metastatic and primary) following 4–6 cycles of chemotherapy to either SBRT to all sites of disease or maintenance chemotherapy. Investigators reported a significant benefit to median PFS (9.7 versus 3.5 months) with the addition of local therapy. Median OS was not reached versus 17 months in favor of local SBRT.⁴ Both trials allowed SBRT to new sites of progression. Notably, these trials were conducted prior to the incorporation of ICIs into first-line systemic therapy for NSCLC. Immune checkpoint inhibitors have significantly improved PFS and OS for patients without driver mutations compared with chemotherapy alone when used in dual-ICI regimens or in combination with cytotoxic chemotherapy. Nevertheless, PFS for these regimens is still measured in months, ranging from 5.1 months in all patients treated with ipilimumab and nivolumab to 11.3 months in patients with PD-L1 expression ≥ 50% receiving pembrolizumab, carboplatin, and pemetrexed.^19–22* Regarding OS, patients with PD-L1 expression ≥ 50% derive the greatest benefit with median OS of 21.9–22.7 months,^22,23 though patients with PD-L1 <1% still benefit with median OS of 17 months.^20,23 While these trials were not restricted to patients with oligometastatic disease, these survival figures are similar or inferior to those in patients treated in the above trials of local consolidative RT and surgery.

A national trial of consolidative local treatment after induction systemic therapy was halted after failing to meet its pre-specified criterion for PFS benefit.²⁴ Though the findings from this trial are not yet published, it may be that incorporating ICI into induction therapy masked the short-term benefit of local treatment, given prior trials were conducted in the pre-ICI era. The SARON and LONESTAR phase III trials are ongoing and will provide additional data on the utility of local therapy for oligometastatic NSCLC.^25,26

Definitive local therapy for metastatic NSCLC is not limited to the upfront setting. The CURB trial randomized metastatic breast cancer and NSCLC patients with 1–5 progressing lesions on systemic treatment to standard of care or SBRT to all sites of progression. While there was no benefit to the addition of SBRT in breast cancer, NSCLC patients (77–82% of whom received immunotherapy) experienced a significant benefit to median PFS (10.0 versus 2.2 months).²⁷ While larger-scale prospective validation is required, the success of this paradigm in NSCLC suggests that local treatments can eliminate pockets of disease that become resistant to systemic treatment, prolonging the effective window of systemic therapies like ICI.

Prior to the introduction of ICI for management of metastatic melanoma, surgery for resection of oligometastatic lesions appeared to improve outcomes for these patients. Studies evaluating surgery for 1–5 metastases identified recurrence-free survival, PFS, and disease-free survival of 5, 7.6, and 8.5 months, respectively, with median OS ranging from 15.8 to 38.6 months.^28–30 Of these, the MMAIT-IV trial was initiated to determine whether a vaccine given in combination with surgery to all sites of disease could improve outcomes. While the vaccine provided no benefit, the 5-year OS rate of 40–45% was notably higher than in other earlier trials of stage IV melanoma. Patients received additional treatments with 44% undergoing surgery following their trial treatments.³⁰ ICI therapy significantly improved outcomes for metastatic melanoma patients with median PFS of current standard of care regimens ranging from 5.5–11.5 months and median OS of 32.7–72.1 months.^31–33 While many patients derive a benefit from ICI therapy, not all patients benefit. In these cases, local therapy to all sites of disease should be considered.

Current NCCN guidelines for metastatic RCC already include local therapy such as SBRT or surgery for oligometastatic disease and oligorecurrent disease, without specifying an upper limit for the number of lesions.¹⁶ A single-arm phase 2 trial of 30 RCC patients with prior nephrectomy and five or fewer lesions treated upfront or following one failed line of therapy with definitive radiotherapy to all sites of disease. Investigators identified a median PFS of 22.7 months. The one-year estimate for systemic therapy free survival was 82%.⁷ Another trial of single-or multi-fraction SBRT for 1–5 metastases from clear cell RCC identified similar outcomes with a median PFS of 15.6 months. Median overall survival was not reached in either trial by the time of publication. Patients were eligible to receive repeat SBRT to new sites of disease.³⁴ By comparison, trials establishing either dual-agent ICI or ICI with a tyrosine kinase inhibitor (TKI) as the current standard for care for metastatic RCC reported median PFS of 11.2–16.6 months with rates of grade 3–5 toxicity of greater than 50%.^35–38 As with lung and melanoma above, it is unknown whether the favorable findings with definitive local therapy could be maintained with treatment of greater numbers of lesions. In principle, offering freedom from systemic therapy to more patients could offer a meaningful improvement in quality of life to patients who could avoid monthly infusions, blood draws, immune-related adverse events (irAEs), and TKI side effects for a year or more. Treatment of oligoprogressive lesions in RCC patients managed with TKIs could also help maintain the benefit of these therapies for longer. A single-arm phase 2 trial of SBRT for 1–5 oligoprogressive RCC lesions treated with TKI alone and identified a 12.6 month median time to change in systemic treatment with a 9.3 mo median PFS.³⁹ These findings support a benefit to local therapy in patients already receiving systemic therapy for RCC.

DEFINITIVE LOCAL THERAPY for DISEASE UNRESPONSIVE TO ICI

For malignancies which do not respond to ICI therapy, local therapy has also been demonstrated to provide a benefit to outcomes across multiple randomized trials. In metastatic prostate cancer, for example, the phase 2 STOMP, ORIOLE, and EXTEND trials identified improved hormone therapy free survival, reduced incidence of progression, and improved PFS, respectively, with the addition of metastasis-directed therapy to patients with oligometastatic prostate cancer following definitive primary treatment.^8–10 Recently presented results of the pancreatic ductal adenocarcinoma arm of the EXTEND trial also demonstrated a PFS benefit of with the addition of definitive local therapy.¹² Another phase 2 trial in patients with oligometastatic (1–4 metastases) esophageal cancer treated with systemic therapy with or without definitive local therapy found a PFS benefit (15 vs 6 months) with local therapy.¹¹ These findings highlight the utility of definitive local therapy in malignancies where ICI is not beneficial.

COMPARATIVE TOXICITY

Though ICIs have a reputation for being well-tolerated, the burden of toxicity imposed by even single-agent checkpoint inhibition can be high. An analysis of 31 trials of pembrolizumab monotherapy identified a 23.7% rate of grade 3–5 irAEs with a treatment-related mortality rate of 0.2%.⁴⁰ Toxicity burden increases with the use of combination checkpoint inhibition. Trials of combination ipilimumab and nivolumab, for example, have reported rates of grade 3 or greater toxicity of up to 48%.^35,41 Although the SABR-COMET trial reported grade 5 toxicity in 3 of 66 patients,² the larger single-arm phase 2 SABR-5 trial which enrolled 381 patients for SBRT treatment of 1–5 metastases reported a 4.2% rate of grade 3–5 toxicity with a 0.3% rate of grade 5 toxicity. Phase 3 studies such as SABR-COMET 3 will help to establish the safety of multi-site SBRT.⁴²

CONCLUSION

Though the concept of delivering definitive local therapy to limited sites of disease is certainly not new, the clinical data supporting this approach is still limited. Further research in this area must accomplish the following:

1. The results of positive phase 2 trials must be reproduced in randomized controlled phase 3 trials comparing local treatment to all sites of disease (alone or with systemic therapy) against standard of care systemic treatment.

2. Trials should investigate treatment of more than 5 sites of disease to determine whether the benefits of local therapy extend to patients with a greater extent of disease. This is ongoing in the SABR-COMET 10 and ARREST trials.^43,44

3. Relatedly, the definition of oligometastatic or oligoprogressive disease must move beyond a count of metastatic lesions. Current and future trials must examine tissue and serum for biomarkers which predict response to local treatment.

4. In malignancies like breast cancer where there is currently no demonstrated survival benefit with definitive local therapy, biomarkers and other clinical and radiologic factors must be investigated to identify subgroups of patients who do benefit.

5. Finally, investigators must assess the optimal timing of definitive local therapy relative to systemic treatment whether up-front, following induction, or at the time of progression. Molecular, histologic, and other characteristics may help inform this choice.

While multiple modalities for definitive local therapy can be used to treat all sites of disease, we emphasize that radiotherapy is well suited for this purpose. Applying artificial intelligence to the radiotherapy planning process will promote faster and more precise treatment of multiple lesions within the same patient while increasing patient throughput by accelerating development of complex treatment plans. Using single-fraction SBRT, which has been demonstrated to be safe and well-tolerated,^45,46 would allow patients to proceed rapidly to systemic therapy if needed. We stress the importance of treating all sites of disease after trials of single-site radiotherapy in combination with ICI in multiple malignancies failed to demonstrate a major survival benefit.^47–50 The purpose of definitive local treatment must be to eliminate all gross disease rather than rely on proposed abscopal effects on the tumor. In patients with limited disease, omission of ICI in favor of local therapy may be possible provided adequate care is taken to surveil these patients for distant spread. This approach leverages the high rates of local control seen with modalities such as surgery and SBRT while sparing patients irAEs.