The prognosis for patients with metastatic non–small cell lung cancer (NSCLC) is poor. However, with recent trials showing improved overall survival with checkpoint inhibitors compared with standard chemotherapy, disease management is changing.
Walter J. Curran Jr., MD
The prognosis for patients with metastatic nonsmall cell lung cancer (NSCLC) is poor. However, with recent trials showing improved overall survival with checkpoint inhibitors compared with standard chemotherapy, disease management is changing. Radiation therapy (RT) is typically reserved for the palliative setting in those patients with metastatic disease because it has been thought to only have effects within the treatment field. With increasing evidence of the immunomodulatory effect of RT, termed the abscopal effect, its systemic effects are being actively investigated. The current understanding of the abscopal effect, and how it can be exploited in those with metastatic NSCLC, will be discussed.
For patients diagnosed with metastatic nonsmall cell lung cancer (NSCLC) without a driver mutation, there has been little therapeutic progress in the past 2 decades.1-3Recent positive results in immunotherapy trials for these patients, however, have begun to transform the management of this disease. As a hallmark of cancer, tumors can develop immune resistance mechanisms via dysregulation of checkpoint protein expression, allowing malignant cells to escape immune destruction.4-6Certain immunotherapeutic approaches allow tumor immunologic tolerance to be overcome and can facilitate a potent anti-tumor T-cell response.7-9A promising class of cancer immunotherapy agents is composed of the immune checkpoint inhibitors, including antibodies against CTLA-4 and the PD-1 pathway. These agents have been found to be active against many tumor types, including melanoma, renal cell carcinoma, NSCLC, and bladder cancer.7,10-12
In NSCLC, recent data suggest those patients with high levels of PD-L1 expression may respond better to checkpoint inhibitors than those with lower levels. Four recent phase III trials have shown that patients with recurrent or stage IV NSCLC and progression after treatment with a platinum-based chemotherapy have improved overall survival (OS) after receiving nivolumab (antiPD-1), pembrolizumab (anti–PD-1), or atezolizumab (anti–PD-L1) rather than docetaxel chemotherapy. These benefits appear related to PD-L1 expression level in the tumor. The checkpoint inhibitors also had a more favorable safety profile than docetaxel.13-16In the first-line setting, randomized results have been less consistent. The CheckMate 026 trial demonstrated no difference in progression-free survival (PFS) or OS outcome between patients with newly diagnosed stage IV NSCLC treated with nivolumab and those treated with platinum-based chemotherapy.17
In contrast, a phase III trial (KEYNOTE-024) showed significantly improved PFS and OS among such patients treated with pembrolizumab compared with physician’s choice platinum-based chemotherapy as first-line therapy; all patients had PD-L1 expression on at least 50% of tumor cells. Again, the immunotherapy arm experienced less toxicity.18Given this result, pembrolizumab monotherapy was approved by the FDA for the treatment of patients with newly diagnosed metastatic NSCLC with PD-L1 expression ≥50%, and pembrolizumab, nivolumab, and atezolizumab are approved as second-line treatments for patients with metastatic NSCLC.
How radiation therapy (RT) fits into this changing landscape remains an open question. For patients with metastatic disease, radiation is typically used for palliation or prevention of symptoms. It is becoming more common at many institutions to offer selected patients with good performance status and low-volume metastatic disease ablative doses of RT to sites of so-called “oligometastatic” disease in the hopes of delaying disease progression and even extending survival.19,20With image-guided RT, along with other technological advances, it is often possible to deliver ablative doses of RT to metastatic foci with minimal damage to surrounding normal tissues. This can be a reasonable treatment option for selected patients with metastatic disease, as a short RT treatment course may minimize the time a patient is off systemic therapy.
RT-induced DNA damage of cells within the irradiated target volume has been considered to be the main mechanism of cell-killing by RT. There is now increasing anecdotal evidence of an RT-induced immune effect. Grills et al published a retrospective review of 124 patients who received either stereotactic body radiation therapy (SBRT) or a wedge resection for stage I NSCLC.21In this study, there was a borderline statistically significant (P = .07) difference in local tumor recurrence rate favoring SBRT. Interestingly, there was also a lower regional nodal recurrence rate in the SBRT group (4% vs 18%). Given that regional nodes were not included in the RT field, this may be an example of an immune response to RT outside the irradiated volume.
While not a new concept, the immunomodulatory effect of radiation has gained traction over recent years and is being actively investigated.22,23First hypothesized by Mole et al in 1953, this has been termed the abscopal effect and refers to a phenomenon in which localized RT leads to tumor response distant from the site of radiation.24-26While the abscopal effect has not been fully elucidated, it is theorized that RT leads to the liberation of tumor antigens and the production of damage-associated molecular patterns, which lead to the maturation of dendritic cells and T-cell priming in an immunogenic context.27Radiation can produce epitope spreading, increasing the diversity of the T-cell receptor repertoire of intratumoral T-cells and broadening the antitumor response.28This can induce and improve an immune response specific to tumor cells.29
RT has also been found to increase T-cell infiltration of the tumor30; the importance of T-cells for a robust response to RT was first described by Slone et al in 1979.31They showed that after immunocompetent and immunocompromised mice with fibrosarcoma were treated with RT, the T-celldeficient mice were less responsive to RT, and they required increasing doses to achieve the same tumor control dose 50% seen in immunocompetent mice. This was subsequently redemonstrated in preclinical and clinical studies.23,32In one of the earliest clinical reports, the abscopal effect was described in a patient with lymphoma.33It was subsequently reported among patients with solid tumors, including NSCLC.34Golden et al described the case of a male with metastatic NSCLC that had progressed through multiple lines of systemic therapy and a definitive dose of RT to his primary thoracic tumor. e developed progressive liver and bone metastases and was started on ipilimumab. RT was delivered to 1 of his hepatic lesions, 30 Gy in 5 fractions, in hopes of triggering both a tumor response and an abscopal effect. The patient had an objective response at all disease sites and had no evidence of disease a year after treatment.35
Although the abscopal effect has been seen when treatment for NSCLC consists of radiation alone, it is thought to occur with greater frequency when radiation is combined with such agents as checkpoint inhibitors.35-37In a subgroup analysis of patients with advanced NSCLC treated on KEYNOTE-001 at a single institution, Shaverdian et al found that patients who received RT prior to their first cycle of pembrolizumab had longer PFS and OS compared with those who did not receive RT. For patients who received extracranial RT, OS was 11.6 months versus 5.3 months, respectively; PFS was 6.3 months versus 2.0 months, respectively. Frequency of grade 3 or greater pulmonary toxicity was not significantly different between those who received thoracic radiation (TRT) and those who did not.38Further, a retrospective cohort study by Hwang and colleagues found that in patients with metastatic lung cancer (95% NSCLC, 5% small cell lung cancer) managed with PD-1/PD-L1 therapy, those who received TRT trended toward improvement in OS compared with those who did not (P = .06). This was despite the fact that patients in the TRT group had fewer targetable mutations. Further, adverse events were similar in the TRT group and no-TRT group.39The sequencing of these treatments is a topic of continued investigation.
Ablative doses of radiation have been associated with an abscopal effect more frequently than standard fractionation.23,40In a study by Lee et al using B16 melanoma-engrafted mice, it was found that ablative doses of RT are more likely to trigger an abscopal reaction than conventional RT fractionation.23Dewan and colleagues showed in preclinical studies that only a hypofractionated regimen of RT was successful at triggering the abscopal effect, while a single-fraction regimen was not, when both regimens were delivered with antiCTLA-4.41In addition, delivering 8 Gy × 3 fractions was found to be more effective than 6 Gy × 5 fractions. Further, a recent meta-analysis found that in preclinical models, the rate of occurrence of the abscopal effect increases with biologically equivalent dose (BED).40Specifically, the probability of triggering the abscopal effect is 50% with a BED of 60 Gy. Despite these intriguing data, however, there is currently no clear consensus on the optimal dose fractionation to trigger an immune response.42
Though previously thought to be primarily a therapy targeting cancer within its treatment field, RT has been shown to have systemic effects. Learning how to exploit these effects remains key to using this modality in a clinically meaningful way for patients with metastatic NSCLC. The combination of RT and checkpoint inhibitors has demonstrated synergy as RT can recruit näive T cells to the antitumor immune response, while antiPD-1 therapy can reinvigorate already primed, but exhausted, tumor-specific T cells.28,43Although this combination has great promise in improving disease-free survival in patients with NSCLC, more work needs to be done to determine which patients can be expected to respond to this therapy. For instance, is the degree of PD-L1 expression predictive of the abscopal effect? What is the optimal RT fractionation? Which metastatic metastatic site should be chosen for focal irradiation? What is the ideal treatment volume? And what is the best timing of delivery of immunotherapy in relation to RT? Several open clinical trials may provide answers to these questions.