Posted: November 12, 2019
Now that data from the PACIFIC trial have been incorporated into daily practice, similar trials are validating the results or taking PACIFIC data in new directions. Greg Durm, MD, is a medical oncologist at the Indiana University Simon Cancer Center and principal investigator of a phase II trial examining concurrent chemoradiation with consolidation pembrolizumab in patients with unresectable stage III NSCLC. In the following interview, Dr. Durm discusses consolidation therapy with immunotherapy or checkpoint inhibitors (CPIs) following chemoradiation.
Q: In the setting of locally advanced NSCLC, do you think consolidation therapy with immunotherapy or CPIs post-chemoradiation is now an established standard?
A: For those patients with unresectable stage III disease, I really do think that consolidation therapy with CPIs postchemoradiation is the standard of care. I realize that there are some patients who will not be candidates for this approach because of concomitant autoimmune problems or other issues. But, for the most part, the bulk of these patients will be able to tolerate these therapies, which are now a key component of therapy after chemoradiation. In patients with resectable disease, the question really becomes whether surgery or consolidation is better. We do not know the answer to that just yet because those trials were done in different eras, and the different modalities certainly have not been compared head to head.
Q: Besides being a straight phase II trial, in what ways did your study of pembrolizumab in this setting differ from the PACIFIC trial?
A: PACIFIC was a randomized trial with a control arm, which did not include any immunotherapy. Ours was a single-arm trial, so all of the patients received pembrolizumab in the consolidation setting. PACIFIC allowed a number of different chemotherapy backbones (with chemotherapy and radiation); ours was more specific in choosing three well-established chemotherapy backbones: carboplatin and paclitaxel, cisplatin and pemetrexed, and cisplatin and etoposide. Lastly, PACIFIC allowed patients to go on consolidation to durvalumab as soon as their physicians felt that they were well enough to do so after concurrent radiation, but our trial required that patients be off therapy for at least 4 weeks. After repeat imaging, patients were able to start consolidation pembrolizumab 4 to 8 weeks after completion of chemoradiation. Other than those few differences, the trials were almost identical.
Q: Do you think there are any fundamental differences between durvalumab and pembrolizumab in the treatment of locally advanced NSCLC?
A: Aside from the obvious difference that one is a PD-1 inhibitor and one is a PD-L1 inhibitor, I think that, based on what we have seen in the metastatic setting both for lung cancer and other tumor types, these behave similarly in terms of efficacy and toxicity. The durvalumab data are just more robust in the stage III setting, given the larger randomized trial.
Q: Are there selection factors (e.g., age and comorbidity status) that influence your decision to adopt this approach?
A: I think the only selection factors that I really take into account are the presence or absence of interstitial lung disease or autoimmune illness, such as rheumatoid arthritis or lupus, or if the patient is on an immunosuppressive agent that I think would lessen the efficacy of the consolidation immunotherapy. I have very successfully treated patients in their late 70s and early 80s, which is a testament to how tolerable these agents are. Otherwise, I consider each patient individually, and I think that most patients—even those with a number of comorbidities—have a really good chance at doing well on these therapies.
Q: Does radiation therapy dose make a difference? What about protons vs photons in this setting?
A: In our trial the patients received definitive-dose chemoradiation, which is typically right around 60 to 66 Gy, which was the same dose range used in the PACIFIC trial. So we know that in the standard definitive-dose range, these medications work. What we do not know is if dose reduction while maintaining efficacy is possible. A number of studies have looked at different doses of chemoradiation, and we know that higher doses above a specific threshold do not work and cause more toxicity. We also know that the 60 to 66 Gy range is effective in treating these patients, even without the addition of CPIs. I think we will likely see the standard of care stay in that dose range, but the answer to whether different doses in conjunction with CPIs work better is largely unknown. The role of proton therapy in lung cancer is still evolving, but there are studies suggesting that it may reduce radiation dose to critical structures. This allows for decreased toxicity but also may improve efficacy by providing safer delivery of dose-escalated therapy to tumor sites and improving the safety and feasibility of multimodality therapy.
Q: Did all patients in the Hoosier Oncology Group trial undergo PET imaging prior to chemoradiation? Is PET prior to chemoradiation routine at this point?
A: Much like the PACIFIC trial, patients in our trial were actually treated with concurrent chemoradiation by their local physicians. Approximately one-third of the patients were treated at Indiana University (IU). The patients did not enroll until after they completed their concurrent chemoradiation and obtained follow-up imaging so we do not have PET data for all 93 patients. PET is routine prior to chemoradiation, which is why I suspect that the vast majority did undergo a PET scan prior to treatment. Before putting someone through chemoradiation and then ultimately a full year of immunotherapy, you really do need to stage them appropriately and ensure that there is no distant disease. In my practice and those of my colleagues here at IU, PET prior to chemoradiation is routine, but I do not repeat PET at 3 months. After I treat patients with concurrent radiation and put them on immunotherapy, I actually just use CT scans to follow them. Of course, if there are questions about whether what we are seeing on the CTs represents actual progression, then we will often order a PET/CT to answer that question. There are emerging data looking at mid-treatment and post-treatment PET/ CTs, but I do not think that this information is ready for general practice yet.
Q: Based on your experience, what percentage of patients with locally advanced NSCLC who go through chemoradiation are ultimately eligible for consolidation immunotherapy?
A: Based on my experience, the vast majority are eligible for consolidation immunotherapy. It’s very tolerable, and the baseline rates of autoimmunity in this population are reasonably low. If a patient is healthy enough to undergo concurrent chemoradiation, that patient is likely healthy enough to get systemic therapy with a single-agent immunotherapy. I would say that the vast majority of patients who go through chemoradiation are eligible for that treatment.
Q: Were there differences in outcome in your trial based on PD-L1 status?
A: Keep in mind that our trial was smaller and did not measure the PD-L1 expression levels for every patient—some of them just did not have enough tissue left over to do that analysis. In our analysis of the HOG trial, the PD-L1 expression level was not correlated with patient benefit from pembrolizumab therapy. PACIFIC offers a much larger dataset, but it is a post hoc analysis. At this point, PD-L1 status is not a factor when I decide to put my patients on consolidation immunotherapy. We know from the metastatic setting that PD-L1 is not the best biomarker. I have personally had a number of patients who had very low PD-L1 expression in the metastatic setting who have responded beautifully to CPIs. I would hate to deprive a patient of the very clear benefit seen in the PACIFIC trial and our trial by relying solely on PD-L1 as the basis for making that decision.
Q: Why do you think previous strategies failed?
A: There is certainly no shortage of trials looking at other strategies. We have tried induction chemotherapy, and all of those studies have failed to improve overall survival (OS). In regard to consolidation therapy, there was a pooled analysis of 41 separate trials, which clearly showed no significant benefit for the addition of consolidation chemotherapy after definitive chemotherapy and XRT. We actually ran our own HOG trial here, with consolidation docetaxel in this setting, and those patients actually did worse numerically with consolidation. We have looked at higher doses of radiation, EGFR inhibitors (both cetuximab and gefitinib), and we have looked at anti-VEGF therapies— all of which have either been proven unsafe (specifically in regard to some of the VEGF inhibitors) or have failed to improve OS in this setting. Why all of those failed when immunotherapy has very clearly succeeded is hard to say. One theory would be that during chemoradiation, patients are receiving chemotherapy, so you probably are selecting out a group of cells that are inherently resistant to those types of treatment. By employing immunotherapy afterward, we not only attack the remaining cancer cells with a different type of strategy, but we also do so at a time when the body is primed to achieve that type of response. As we know from a number of preclinical models and now from clinical trials, radiotherapy seems to sensitize the body to immunotherapy. It increases release of neoantigens, it increases tumor immune cells at or around the tumor bed, and it decreases some of the immunosuppressive effects of the tumor microenvironment. So there are a lot of reasons why immunotherapy may be successful in this setting. It may be that we have found the appropriate timing for these or that we have found a different modality that works in a different way than chemotherapy. Either way, I do not know that there is a clear answer to that question, but these are my theories.
Q: What fundamental questions regarding CPIs in this setting remain?
A: What is the proper timing? In PACIFIC, the patients were able to go on immunotherapy directly after chemoradiation therapy was completed. Some of those patients started treatment in the first couple of weeks; in our trial they were enrolled a little bit later, as was previously mentioned. Both obviously showed improvements in progression-free survival (PFS). PACIFIC showed improved OS; there also was a post-hoc analysis examining earlier vs later start for immunotherapy and suggested that those patients who started earlier may have done a bit better. In our trial, we did a similar analysis, again with a smaller number of patients, and we did not see much of a difference for those patients who started earlier in the course (Weeks 4 to 6) versus later in the course (Weeks 6 to 8). I think the danger of making that assumption is that patients who start early obviously did very well with chemoradiation; these patients may be a different patient population, however, compared to those who might have struggled through treatment. Attributing all of the benefit to the early administration of immunotherapy is a little bit difficult.
What is the proper duration? PACIFIC chose 12 months of immunotherapy; our trial also chose 12 months of pembrolizumab. Is 12 months enough treatment? Clearly there was benefit, but would 24 months show more? Or would 6 months perhaps be enough? These drugs are typically well tolerated but can demonstrate significant toxicities as well. We did see some increase in immune-related toxicities in both our trial and in the PACIFIC trial. We also know that these drugs are costly, so I think that future trials need to address whether shorter periods are just as effective, or if 12 months is the optimal duration for these patients.
Would combination therapies improve results? Using a PD-1/PD-L1 inhibitor in combination with a CTLA4 inhibitor is one strategy, and there are a number of other CPIs in development at the current time. Furthermore, there are ongoing trials looking at the addition of immunotherapy in combination with chemoradiation, which is another interesting strategy.
What role does surgery play in all of this? We do know that both our trial and PACIFIC were conducted in patients with unresectable disease, but approximately 20% of our patients have resectable stage III disease. Whether it is more beneficial for them to undergo surgery after chemoradiation or to begin consolidation immunotherapy is largely unanswered. Perhaps, in the future, it will be some combination of those two strategies, and there is at least one ongoing trial looking at incorporating both strategies. ✦