During a <em>Targeted Oncology </em>case-based peer perspectives program, Ryan Gentzler, MD, reviewed with other physicians his clinical considerations for the management of non–small cell lung cancer.
Ryan Gentzler, MD
During aTargeted Oncologycase-based peer perspectives program, Ryan Gentzler, MD, reviewed with other physicians his clinical considerations for the management of nonsmall cell lung cancer (NSCLC). Levy, assistant professor, department of hematology and oncology, University of Virginia Health System, discussed his treatment options and other factors that go into his decision making with the group during the meeting based on a case scenario of a patient with unresectable locally advanced lung adenocarcinoma.
A 63-year-old man presented to his primary care physician with intermittent cough and difficulty breathing on exertion. His past medical history showed hyperlipidemia, which was well managed on simvastatin (Zocor); hypothyroidism, managed on levothyroxine (Synthroid); and chronic obstructive pulmonary disease, which was managed on inhalers. He recently quit smoking but had a history of 40 packs per year. His physical exam showed intermittent wheezing and his ECOG performance status score was a 1. His lab value for creatinine clearance was within normal limits.
A chest CT revealed a 3.1-cm spiculated mass in the right upper lobe and 2 enlarged right mediastinal lymph nodes, 1 measuring 1.5 cm and 1 measuring 1.7 cm; moderate emphysema was noted. A PET scan confirmed the lung lesion and mediastinal lymphadenopathy without evidence of distant metastases. His brain MRI was negative, and a pulmonary function test revealed forced expiratory volume 1, 1.2; diffusing capacity of the lungs for carbon monoxide, 35%.
A bronchoscopy with transbronchial lung biopsy and lymph node sampling revealed adenocarcinoma with positive nodes in stations 4R and 7; level 4L was negative. His staging was T2aN2M0, stage IIIa. Based on the extent of mediastinal disease and emphysema, the patient’s cancer was deemed inoperable, and he was referred for consideration of concurrent chemotherapy and radiation.
Targeted Oncology:Do you typically order genetic testing for locally advanced lung cancer?
Gentzler:As of 2018,EGFRmutations,ALKrearrangements, and even PD-L1 status don’t necessarily play into the decision making for stage III lung cancer. We do it [at our practice] as a routine because there’s a high chance these patients will progress at some point and develop metastatic disease. As long as we have adequate tissue, we feel it’s worthwhile to do it so that if and when they do progress, we are ready to go with whatever the most appropriate treatment is. But it’s not necessarily standard. You could, at the time of progression or at the development of metastatic disease, go back and analyze the archival tissue or get a new biopsy and do that testing. I think there is some variation around the country regarding when to do molecular testing.
What are the treatment options for this patient?
We could give this patient neoadjuvant chemotherapy, neoadjuvant chemoradiation, concurrent chemoradiation, or sequential chemotherapy followed by radiation. In this case, I would give chemotherapy and concurrent radiation.
Multiple studies over the years have looked at this question, and there is no benefit to induction or neoadjuvant chemotherapy followed by chemoradiation. We often do sequential chemotherapy and radiation for patients who are less fit and might not be able to tolerate concurrent therapy, but this has been shown to be inferior to concurrent therapy.
You could pursue any of these approaches followed by surgery, and I think that this is the controversy in someone who is fit for surgery. We have data that neoadjuvant chemoradiation followed by surgery may improve survival compared with definitive chemotherapy and radiation alone. That is why we discuss patient cases at tumor boards and get a sense from our surgical colleagues of who is resectable and who’s not, what the size of the lymph nodes are, and if this is going to be a difficult case to get a negative margin. There’s a lot of nuance to that.
He underwent therapy with cisplatin/etoposide and concurrent thoracic radiotherapy. Follow-up imaging showed a partial response with shrinkage of the primary and nodal lesions.
How would you follow this patient after concurrent chemoradiation?
The patient should have a CT scan every 3 months, which is standard. There have never been any trials looking at how often to do surveillance in these patients. We typically do 3 months for stage III patients who did not have surgery.
What is the historical survival rate with chemotherapy and radiation for stage III NSCLC?
In the PROCLAIM trial, [investigators] looked at 2 chemotherapy regimens for patients with adenocarcinoma. [The patients] were all given radiation, and the median survival numbers were between 25 and 26.8 months.1RTOG 0617 is another trial that used different chemotherapy regimens for patients with adenocarcinoma and squamous NSCLC.2It was a more heterogenous NSCLC population. In this study, the median overall survival (OS) was 28.7 months, and the 5-year OS was 32%. The odds are that this cancer is going to come back at some point. It is a high-risk situation.
Does this patient require further treatment after achieving a partial response?
The PACIFIC trial, a recent trial presented at the European Society for Medical Oncology Congress, looked at this question. It was a phase III, randomized, double-blind, placebo-controlled design.3The patients in this study had stage III locally advanced, unresectable NSCLC. They had to complete definitive chemotherapy and platinum-based radiation and had to receive at least 2 cycles of chemotherapy. They did collect archival tissue, but tissue biopsies were not required for study entry.
These patients were randomized 2:1 to durvalumab (Imfinzi) or placebo. Durvalumab was given 10 mg every 2 weeks for up to a year. Progression-free survival (PFS) and OS were both the primary endpoints. Looking at the PFS curves, there a clear benefit with durvalumab over placebo. The median PFS in the placebo arm was 5.6 months versus a substantial 16.8 months in the durvalumabarm.
Did a specific subset of patients benefit more from this study?
When you break it down by different demographics, such as PD-L1 andEGFRstatus, the forest plot hazard ratios (HRs) were on the left side of 1 favoring durvalumab, with a few exceptions.
We were all interested in the PD-L1 status. Investigators used different cutoffs than what we use for other clinical scenarios, but if the patient had low PD-L1, high PD-L1, or unknown, all 3 of those seemed to benefit from durvalumab. If you look at patients with stage IIIa or IIIb, or smoker versus nonsmoker, it didn’t seem to make much of a difference.
One thing that came out of another analysis in the supplemental index of this publication was the question of how quickly they were randomized. If you look at the difference between patients who were randomized before or after 14 days of their last radiation treatment, the HRs are 0.39 versus 0.63, respectively.
There are 2 theories as to why this could be. It could be that there isn’t a difference and that these are still overlapping confidence intervals with small numbers. It also could be that patients who do well with chemotherapy and radiation are more fit and will have a better prognosis regardless of what the treatment is. There’s also a thought that, in the metastatic setting, the combination of chemotherapy plus immunotherapy seems to have benefit. It could be that the closer you get the immunotherapy to the chemotherapy and radiation, the more there is the potential for synergistic benefit. So, I think this idea here is driving future trials for potential of synergistic benefit.
I think these ideas are driving future trials to start durvalumab earlier in combination with chemotherapy and radiation or before chemotherapy and radiation. I think it’s worth pointing out. This has also changed our practice. We used to scan a month after chemotherapy and radiation, and now we’re trying to scan the patients a week or 2 within finishing radiation and then move right onto durvalumab if they qualify for that treatment.
What is the significance of the updated data from the PACIFIC trial?
Just a couple of weeks ago, we received the update on this trial at the International Association for the Study of Lung Cancer (IASLC) World Conference on Lung Cancer.4The control arm had a median OS of 28.7 months. Whenever you look at a trial, you want to question how the control arm faired. Was it a particularly sick group of patients compared with historical controls, and the experimental outperformed because the other arm underperformed? However, we see that this matches the 0617 trial exactly. In this case, this is a fair assessment of our current standardized care. For OS, we see a clear benefit with durvalumab. The curves separate early, and at each landmark, there is a clean benefit. We haven’t reached a median OS with durvalumab, and they are still censoring these patients after 2 years. This is impressive and confirms what we suspected based on the PFS data alone.
They also looked at another endpoint called time to distant metastasis or death, which is a surrogate for metastatic disease. When you look at the data, you can see that these numbers are about a year shorter than when these patients eventually died. Additionally, they went back and looked at PD-L1 status at different cutoff points. A lot of us wondered why there was a 25% cutoff and questioned how the truly negative patient, or less than 1%, performed. When you look at the PFS, it seems that the HR crosses 1 in the less than 1% patient population. I don’t think we know how to interpret these data. Additionally, not all these patients had tissue that was evaluable for PD-L1 status. It’s hard to draw conclusions, but it makes you wonder if the patients who had a PD-L1 status of less than 0% are really benefiting.
What was the safety profile of durvalumab in the PACIFIC trial?
In terms of safety, durvalumab performs similar to other PD-1 or PD-L1 drugs. You see grade 3/4 adverse events (AEs) in the 20% to 30% range. There were grade 5 AEs. As a reminder, grade 5 AEs means death. But, we saw that in the placebo arm as well with a similar rate. I think this tells us that the effects of chemotherapy and radiation are not benign. Additionally, in the study there were not a whole lot of differences in any of the toxicity [between the 2 arms], with the exception of the immune-related AEs (irAEs), with a rate of 24% in the durvalumab arm and 8% in the placebo arm. Remember, this is placebo controlled, and the investigators didn’t know what they were giving. If they felt that a patient had an irAE, they categorized it as that.
Pneumonitis was a large concern before this trial, and there is typically a lot of hesitation not to overlap a lot of thoracic radiation with immunotherapy. The potential for pneumonitis was certainly feared. When you look at grade 3/4 pneumonitis in the study, the rate was fairly small at just 3.4%. The any-grade AE of pneumonitis was 34%, and this includes radiation pneumonitis or drug-induced pneumonitis, which is difficult to tease out in this population. To conclude, grade 3/4 pneumonitis was similar in both the durvalumab arm and the placebo arm, and for any-grade pneumonitis, it was higher in the durvalumab arm.
Would you consider additional options for this patient after concurrent chemoradiation?
Aside from durvalumab, there are different studies that asked this question. There was a phase II trial done by the Hoosier Cancer Research Network led by Greg Durm.5It had a very similar design. Patients were allowed to get any of the standard platinum doublet treatments with the addition of radiation 59.4 Gy to 66.6 Gy. They had repeat CT scans, which was mandated a little bit later in this trial, so it had to be a month to 2 months after. Those with progressive disease were not eligible, and those with stable responsive disease went on to get pembrolizumab (Keytruda), which is another PD-1 inhibitor, every 3 weeks for up to 12 months.
The median time to metastatic disease was 30.7 months. The median OS was not yet reached. If we look at the OS curve on this trial compared with the durvalumab arm on the PACIFIC trial, we can see fairly similar results. It’s nice to see a different trial with a different drug coming up with the same result. It substantiates the results of the PACIFIC trial that this is an effective strategy and has clear benefit in terms of OS.
Is PD-L1 status a substantial biomarkerfor stage III NSCLC?
We don’t know the answer to that. We think that, unlike genetic mutations that persist throughout the course of chemotherapy treatment, PD-L1 is an inducible biomarker. In the setting of chemotherapy, radiation, or immune response, PD-L1 can be more variable. It’s also a more subjective biomarker. You need to get a biopsy of the tumor, and it depends where in the tumor you get the biopsy, because there is heterogeneity within the tumor. There is also some subjectivity when reading the test. There is a score, and you can count cells to make an assessment. Although I think people have been fairly consistent in their scoring, it does require a pathologist to visually look at a slide and make that determination.
There’s a lot of controversy surrounding PD-1. The different assays and antibodies have all been tested in different studies. So, there are caveats with PD-L1 testing, and looking at the subgroup analysis of a PD-L1 result that was from archival tissue before the chemotherapy and radiation may or may not be the best [approach].
In a large phase I trial of pembrolizumab in patients with stage IV disease, [investigators] looked a little bit at this question and the length of time that lapsed from when a patient had a biopsy to the time they were tested for PD-L1. It didn’t seem to matter whether it was a fresh biopsy or 1 that was archived from a year ago. The PD-L1 status didn’t seem to be affected by time, but that does not necessarily answer this question. It is still unknown. Additionally, no studies have looked at serial PD-L1 testing throughout the course of treatment. It is something worth looking at in future studies.
Would you consider neoadjuvant immunotherapy for patients with stage III NSCLC?
We have data in lung cancer for nonadjunct immunotherapy with a couple of different designs with very impressive results. In theNew England Journal of Medicine, a small series run by Johns Hopkins University School of Medicine and Memorial Sloan Kettering Cancer Center looked at 2 doses of single-agent nivolumab (Opdivo) followed by surgery for surgically resected, earlier-stage lung cancer.6The radiographic response was about what we expect; 20% to 30% of patients had a response. However, when they looked at pathological response, that number was substantially higher, so we don’t know how that portends to OS. This was a good group of patients with early-stage disease, and it will take a long time to follow to see if they have disease recurrence.
We also saw some data at IASLC from a Spanish trial that looked at neoadjuvant chemotherapy plus radiation prior to surgery for surgically resectable patients.7This was for patients with stage III lung cancer, and almost all the patients had a significant pathological response. [The investigators] reported a 75% complete response pathologically, which is much better than anything we’ve ever seen before with neoadjuvant chemotherapy. That is another area that’s actively being investigated, and it speaks to this chemoradiation question and matching up the immunotherapy closer to the treatment. If you give a couple of doses of immunotherapy followed by chemotherapy and radiation, are you going to get a better response? That is a question that still needs to be answered.