Checkpoint Inhibitors and Clinical Decision Making: A Q&A with Dr. Nasser H. Hanna

Dr. Nasser H. Hanna

Posted: May 27, 2019

Nasser H. Hanna, MD, is the Tom and Julie Wood Family Foundation Professor of Lung Cancer Clinical Research at Indiana University School of Medicine, and he specializes in the study and management of all forms of lung cancer. Dr. Hanna spoke with the IASLC Lung Cancer News in detail about the questions— asked and remaining—surrounding the use of checkpoint inhibitors (CPIs) in the front-line setting for squamous and nonsquamous NSCLC. In the interview below, Dr. Hanna shares information about his personal clinical decision-making process, and he highlights studies—published and ongoing—that provide a solid roadmap for therapeutic selection in each course of treatment.

Q: As CPIs move to the front line, which regimen or regimens are now “standard” in the second-line setting in advanced NSCLC?
A: It is now standard of care to treat all patients with squamous cell NSCLC who are chemotherapy naive but are eligible for pembrolizumab with carboplatin and either paclitaxel or nab-paclitaxel plus pembrolizumab. Although many patients benefit from this combination, just about everyone will experience disease progression, which leads to the question of what to do in the second line for these patients. The standard-of-care treatment for nearly 2 decades in the second line for squamous cell has been docetaxel, which will continue to be an option for those patients who can tolerate treatment once every 3 weeks (usually at 60-75 mg/m2 or once weekly, 3 weeks on/1 week off at 35 mg/m2). Alternatively, patients can receive combination docetaxel/ramucirumab based on the results of the REVEL study.1 This is very viable option for patients who are more fit and who do not have contradictions to ramucirumab, which is an antibody to the VEGF receptor. For nonsquamous NSCLC, patients who have a PD-L1 tumor progression score (TPS) of at least 50% are eligible to receive pembrolizumab in the first line, and many of those patients will benefit for a time and then experience disease progression. For these patients, I think the standard of care will be carboplatin plus pemetrexed with or without bevacizumab. For patients with nonsquamous NSCLC who receive carboplatin plus pemetrexed plus pembrolizumab in the first-line setting and then subsequently experience disease progression, I think that, once again, docetaxel given once every 3 weeks, docetaxel given weekly for 3 weeks on/1 week off , and docetaxel plus ramucirumab are all options.

For both squamous and nonsquamous histologies, some patients will still be treated with gemcitabine, and other patients will go on clinical trials. There will be some patients who have very slow minimal progression who will continue to receive pembrolizumab for a bit longer in the first-line setting, and some patients with oligometastatic progression may receive local therapy and continue on first-line therapy.

The question that I think a lot of people want to know the answer to is, if a patient receives chemotherapy plus a CPI in the first-line setting, is there any value in continuing the CPI at the time of progression? In those patients who have slow progression, there may be some value in continuing the CPI. Response data has shown that there are clear times of tumor growth and shrinkage. Based on this, I tend to continue the CPI if the patient is doing clinically well, even if there is some temporary disease progression. There are a number of ongoing trials examining the mechanism of resistance to CPIs and the potential benefit of CPIs in combination with chemotherapy.

Fortunately, toxicities usually present themselves early in treatment; typically, the patients who get beyond 3 or 4 months of treatment with a CPI tend not to have any major toxicities thereafter. I would not anticipate that re-challenging someone with a CPI or switching to a different CPI is likely to cause any change in that side-effect profile.

Q: Does this reinvigorate the role of combination docetaxel and ramucirumab in this setting?
A: Docetaxel plus ramucirumab has demonstrated a survival advantage over docetaxel alone, which is not a small bar to eclipse. I think this combination is a very reasonable and acceptable option for many patients in the second-line setting who have already received a CPI with chemotherapy in the first-line setting regardless of histology (squamous and nonsquamous) and even for those patients who had previously received a taxane in the first-line setting. Previous to the CPIs moving into the first-line setting, the standard was typically chemotherapy in the first-line setting and single-agent CPI in the second-line setting, and then docetaxel with or without ramucirumab in the third-line. As you go from first to second line and second to third line, fewer and fewer patients are likely to get those treatments for a variety of reasons; therefore, anytime a therapy moves from the third to the second line it means, by definition, that more patients are likely to be getting that regimen.

Q: With respect to patients who have been on chemotherapy/CPI combinations front line, do you think there is a role for platinum re-challenge in patient who have stabilized or responded to prior platinum regimen(s) and who experience disease progression on CPI alone or on pembrolizumab/ pemetrexed?
A: I think that in a data-free zone, we use the principles from other settings and we extrapolate information. Generally speaking, if patients have responded to prior platinum-based chemotherapy and it has been more than 6 months—certainly more than 1 year—since they have been exposed to a platinum agent and the patient continues to have a good performance status and is appropriate for combination therapy, I think it is very reasonable to add a platinum agent back. I do not think it makes any sense to do so for a patient with a poor performance status or for a patient who did not seem to have a long duration of benefit from a platinum-based therapy in the initial setting. This is a general principle that we have used for years, particularly for patients with SCLC who have been off a platinum regimen for more than 6 months and then experience disease progression.

Q: In a patient with nonsquamous NSCLC, would you consider resuming carboplatin, substituting a taxane for pemetrexed, and switching the CPI for an angiogenesis inhibitor? Why or why not?
A: That’s a clinical judgement. There are now data from IMpower 150, which is a randomized phase III trial in the firstline setting of patients with nonsquamous NSCLC who were randomly assigned to receive carboplatin plus paclitaxel plus atezolizumab plus bevacizumab versus a comparator arm of carboplatin plus paclitaxel plus bevicizumab.2 Th is study demonstrated improved PFS and OS for the four-drug regimen (HR = 0.775; 95% CI: 0.619-0.970; p = 0.0262), which proved both statistically and clinically significant. This regimen was approved by the U.S. Food and Drug Administration on December 6, 2018. Th is really does address several questions: whether we can go back to a platinum-based regimen for patients who have been off of platinum-based chemotherapy for a while, if administering a taxane to a patient in the second-line setting who previously benefited from pemetrexed is worthwhile, and if adding an angiogenesis inhibitor in the second- line setting like we do with ramucirumab is worthwhile. I think those individual answers are all potentially “yes”; now, whether collectively they are appropriate for patients really remains to be seen and requires a great deal of clinical judgment. Again the individual principles hold up: re-challenge with a platinum, switch to a taxane, and add an angiogenesis inhibitor.

Q: What role is there for adding new immunotherapeutic agents (e.g., vaccines and CTLA-4 inhibitors) to frontline CPIs in those with “smoldering progression”?
A: This presumes that a patient received either single-agent pembrolizumab or chemotherapy plus pembrolizumab and then experiences disease progression. What is the value of maintaining that and simply adding on some other immunotherapy strategy, whether a vaccine or new class of immunotherapy, such as a CTLA-4 inhibitor? I think at this point, we have to just sit back and wait for the results of clinical trials. I do not know that there are any early data that are overly promising for that strategy with our current drugs. Vaccines are still significantly behind other immunooncology drugs with respect to lung cancer. I think that some patients will begin to receive combination immunotherapy in the first-line setting. There are data from the CheckMate 227 trial regarding nivolumab plus ipilimumab in comparison with chemotherapy, single-agent CPI, or chemotherapy/ immunotherapy for patients with high tumor mutation burden (TMB), and it does appear that the duration of response and the PFS may be better with that combination in the first-line setting.3 Now whether adding ipilimumab, tremelimumab, or some other CTLA-4 inhibitor to a CPI after the patient has already experienced disease progression on the CPI has any benefit is purely speculative; we really must await results of clinical trials.

Q: How do you handle “oligoprogression,” where most sites are stable or responding but one or more are growing? Do you pursue locally ablative therapy and continue the original CPI? Or do you switch regimens?
A: This is a wonderful question because we have come to understand that different metastatic sites have different biologic, make-up and that in turn can explain why disease can be responsive in one area and progress in another area. There are a number of studies in lung and other cancers in which biopsies of different metastatic sites or a biopsy of a single metastatic site are compared with the primary site, demonstrating a different molecular biology for each site. We do this with patients who have oncogenic- driven cancers, such as EGFR, where patients can continue to have excellent therapeutic responses generally but experience disease progression in a few areas. We have tended to treat those areas locally and maintain the TKI, but I think a similar strategy can be looked at with the CPIs. One cautionary note about overinterpreting progression with CPIs is that sometimes patients do experience psuedoprogression, and sometimes patient responses wax and wane. We see this in the spider plots, which show from cycle to cycle how much the disease burden has increased or decreased; and we know that there is some fluctuation with immunotherapy. For many patients, treatment continues as long as they are clinically well. If it is clear that disease has progressed at a few sites but the patient is doing well otherwise and the majority of the disease burden is well controlled, I use stereotactic radiation to treat the local progression.

Q: How long are you treating patients with CPI in the second line? Does it differ based on drug?
A: I am treating patients until disease is clearly progressing or until the patient has a toxicity that requires them to discontinue treatment. We are awaiting the results of clinical trials to understand the optimal duration of therapy. With the original study of nivolumab, patients were randomly assigned after 1 year of therapy if they did not have disease progression; this was generally with nivolumab given in the second-, third-, and fourth-line setting—not in the first-line setting.4 After 1 year, that study demonstrated that the patients who stopped the nivolumab had more rapid disease progression; PFS favored those who continued. Although a study like this was not designed or powered to show OS, the trend did favor continuation of the nivolumab. This was true for patients who had a partial or complete response to nivolumab, as well as for patients whose best response was stable disease. The studies with pembrolizumab, generally in the second-line setting, allowed for up to 2 years of therapy.5 There are trials that are ongoing, particularly in Europe, that are looking at increasing the interval between treatments once you get to 1 or 2 years out, so patients might be randomly assigned to receive a drug every 3 weeks versus every 2.6 Clinical trials are going to try to address this, but with the current data that we have, we do continue to treat as long as the patient appears to be clinically benefitting without experiencing a toxicity that requires discontinuation.

Q: Do some biomarkers (TMB for example) infl uence your choices to (re) use immunotherapy based on a patient’s PD-L1 expression level?
A: The issue of biomarker development for immuno-oncology is critically important. It will help us identify patients who are destined not to benefit, as well as patients who are more likely to benefit from maybe more aggressive approaches such as combination therapy. Ultimately, our patients do want to be those 3-, 4-, and 5-year survivors and beyond, so we do want to optimize treatment; biomarkers have gone a long way in helping us do that. In fact, I would argue that we have better biomarkers for immuno-oncology than we ever had for chemotherapy. Some of those biomarkers include: PD-L1 expression; TMB; single mutations (such as STK11 mutations, which seem to confer potential resistance); other parameters, such as neutrophil-to-lymphocyte ratios; presence of tumor-infiltrating lymphocytes; CD4 to CD8 ratios; and understanding the polymorphisms of the T cell receptor. There are a collection of clinical and biologic biomarkers that I think are all very interesting and which we are already beginning to use to some extent. In terms of PD-L1 expression specifically driving a decision in the second-line setting, I do not really think that is the case. I think it may play a larger role in something like whether we give nivolumab and ipilimumab without chemotherapy in the first-line setting rather than something like single-agent pembrolizumab or chemotherapy plus pembrolizumab. I think this will all be much clearer in the next year or two.

Q: Are you involved in any ongoing research regarding immunotherapy?
A: We have two studies that we are conducting in the Hoosier Oncology Group, both of which are investigator-initiated trials. One of these studies is in the socalled “immunoresistant” population and one is in the so-called “immunosensitive” population. We have defined “immunoresistant” patients as those who had prior platinum-based chemotherapy and a CPI but have experienced disease progression within three treatments of the CPI (NCT 02343952). These patients are being randomly assigned to singleagent chemotherapy versus chemotherapy/ CPI as a way to determine whether there is any value of CPIs for patients who did not seem to benefit from CPIs in the first-line, but who perhaps could be re-sensitized or sensitized in the first place to the CPI simply by adding a different chemotherapy agent to it. “Immunosensitive” patients are those who have responded for a minimum of 3 months to a previous CPI (NCT 03083808). Our trial continues the CPI and adds a different chemotherapy agent to it at the time of progression. Again, this trial is really addressing the same question but in a different patient population. The idea behind this is that chemotherapy plus a CPI may be more effective than the CPI alone because the chemotherapy can cause some cellular kill and can introduce new neoantigens into the tumor microenvironment, thereby making that microenvironment more hospitable to the immune system, with improved capacity to recognize cancer antigens and mount a more robust response. One of our studies is approximately halfway complete, but it will take a couple of years.

The positive aspect to remember about the immunosensitive population is that when a patient responds to a singleagent CPI, the duration of response tends to be quite long—it is not usually measured in months but in years. This, however, requires a longer trial-recruitment period. For the immunoresistant population, these patients also tend to be chemotherapy resistant, so there are a number of complicating issues with trial enrollment. These trials are challenging to conduct. In theory, it is wonderful to just write down on a notepad these different trial designs, but in reality it is challenging to keep track, and it takes a considerable amount of time. ✦

1. Garon EB, Ciuleanu TE, Arrieta O, et al. Ramucirumab plus docetaxel versus placebo plus docetaxel for second-line treatment of stage IV non-small-cell lung cancer after disease progression on platinum-based therapy (REVEL): a multicentre, double-blind, randomised phase 3 trial. Lancet. 2014;384(9944):665-673.

2. Socinski M, Jotte RM, Cappuzzo F, et al. Overall survival (OS) analysis of IMpower150, a randomized Ph 3 study of atezolizumab (atezo) + chemotherapy (chemo) ± bevacizumab (bev) vs chemo + bev in 1L nonsquamous (NSQ) NSCLC. J Clin Oncol. 2018;36(suppl; abstr 9002).

3. Hellmann MD, Ciuleanu TE, Pluzanski A, et al. Nivolumab plus Ipilimumab in Lung Cancer with a High Tumor Mutational Burden. N Engl J Med. 2018;378(22):2093-2104.

4. Spigel DR, McLeod M, Hussein MA, et al. Randomized results of fixed-duration (1-yr) vs continuous nivolumab in patients (pts) with advanced non-small cell lung cancer (NSCLC). Annals of Oncology (2017) 28 (suppl_5): v460- v496.

5. Herbst RS, Baas P, Kim DW, et al. Pembrolizumab versus docetaxel for previously treated, PD-L1- positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomized controlled trial. Lancet. 2016;387(10027):1540-1550.

6. Herbst RS, Garon EB, Kim DW, et al. Long-term follow-up in the KEYNOTE-010 study of pembrolizumab (pembro) for advanced NSCLC, including in patients (pts) who completed 2 years of pembro and pts who received a second course of pembro. Annals Oncol. 2018;29(suppl_10):LBA4.