In a Targeted Oncology case-based peer perspectives live discussion with a group of physicians, Corey J. Langer, MD, reviewed the potential for treatment with immunotherapy following chemoradiotherapy for patients with stage III non–small cell lung cancer. Langer discussed these options based off of a case scenario of a patient with stage IIIA lung adenocarcinoma.
Corey J. Langer, MD
In aTargeted Oncologycase-based peer perspectives live discussion with a group of physicians, Corey J. Langer, MD, reviewed the potential for treatment with immunotherapy following chemoradiotherapy for patients with stage III nonsmall cell lung cancer (NSCLC). Langer, a professor of medicine, Hematology/ Oncology Division, Hospital of the University of Pennsylvania, and director of thoracic oncology, Abramson Cancer Center, discussed these options based off of a case scenario of a patient with stage IIIA lung adenocarcinoma.
A 63-year-old man presented to his primary care physician with intermittent cough and difficulty breathing on exertion. Past medical history was remarkable for hyperlipidemia, which was well managed on simvastatin; hypothyroidism, which was managed on levothyroxine; and chronic obstructive pulmonary disease, which was managed on inhalers. He had recently quit smoking but had a 40-pack-year history.
On physical exam he showed signs of intermittent wheezing and had an ECOG performance status of 1. His complete blood count, chemistries, and creatinine bloodwork were all within normal limits.
A chest CT revealed a3.1-cm spiculated right upper lung mass, 2 enlarged right mediastinal lymph nodes measuring 2.5 cm and 1.7 cm, and moderate emphysema. A PET scanconfirmed the lung lesion and mediastinal lymphadenopathy without evidence of distant metastases. Brain MRI came back negative.
His FEV1 measured 1.2 and his DLCO was 35%. A bronchoscopy with transbronchial lung biopsy and lymph node sampling revealed adenocarcinoma at primary site with positive nodes in stations 4R and 7; level 4L was negative. The patient was diagnosed with NSCLC T2aN2M0, stage IIIA.
Based on the bulk and extent of mediastinal disease and active emphysema, the patient’s cancer was deemed unresectable, and he was referred for consideration of concurrent chemotherapy and radiation.
Targeted Oncology: Do you get PET scans for everyone?
Langer:Almost everyone, with 2 exceptions. First, patients with flagrant metastatic disease where I’m not sure it’s going to add anything. They present with liver and bone metastases, so what’s the point?although inevitably you find something that probably wasn’t on the scan. Second, if it’s a too-tiny solitary pulmonary nodule. Even then, we’ll probably do it, but the argu­ment is that the CT is negative for nodal enlargement, and our surgeon will often go in and convert a biopsy to a lobectomy, and the node dissection. Yes for a PET scan on pretty much everyone else.
For [a patient with stage] IIIB disease, I’d fight [for a PET scan] because about 20% of the time you’ll find metastatic disease in IIIB and an aggressive combined modality approach probably wouldn’t be appropriate. In part, I’d do it [for stage III patients] so the bron­choscopist can figure out which is the hottest node to go after. Frankly, that may be 1 reason we see such a huge survival advantage [in the PACIFIC trial]. Remember, the clock starts ticking after the chemoradiation; PET wasn’t necessarily done routinely in that study.
Targeted Oncology: Do you typically order genetic testing for this type of patient?
Langer:Officially we don’t.
Targeted Oncology: What are the options for treatment for this patient with locally advanced nonsmall cell lung cancer (NSCLC)? What chemotherapy regimen would you recommend if chemotherapy is given?
Langer:I don’t think this person would get surgery, not with those PFTs and not after chemoradiation. If the patient were stage IIIA, some of my colleagues would still like to use cisplatin/etoposide; I pretty much use paclitaxel/carboplatin exclusively. Any of the [National Comprehensive Cancer Network (NCCN)recommended] chemotherapy regimens can be covered.1So take your pick.
Targeted Oncology: Would you lower the radiotherapy dose?
Langer:Yes, to 60 Gy. I wouldn’t go lower; there’s no justification. The RTOG 0617 study was a negative state-of-the-art trial; it was a 2:2 design looking at radiotherapy and paclitaxel/carboplatin, either high-dose radiotherapy or higher-dose radiotherapy (60 Gy vs 75 Gy) plus or minus cetuximab [Erbitux].2It was negative on the radiation push (75 Gy); they had a statistically significant survival detriment. It was thought to be in part due to the heart and other comorbidities that seemed to be activated. With 60 Gy plus weekly paclitaxel/carboplatin, the median survival was 28 months. That’s our benchmark these days.
CASE (continued):patient underwent therapy with cisplatin/etoposide and concurrent thoracic radiotherapy. Follow-up imaging showed a partial response with shrinkage of the primary and nodal lesions. No evidence of distant spread was observed.
Targeted Oncology: How do you follow this patient?
Langer:Nobody has ever established that; it’s just been baked into all of the [insurance]. Usually [insurance] will pay for every 3 months; getting [scans] more often is probably unnecessary. I have some colleagues who do it every 6 months. It’s all about how you got trained, and it’s also based in part on how clinical trials were done.
Targeted Oncology: Could this patient benefit from further treatment? What data support your decision?
Langer:I was highly skeptical of the PACIFIC trial when it was first started, but it was a stroke of genius to look at this drug in an earlier-stage setting. [The trial had a] 2:1 randomization of durvalumab [Imfinzi] versus placeboa big, global study of 700 patients.3Chemoradiation was given upfront, and someabout 15% to 20% or so—got induction as well, in addition to chemoradiation. [They had to have a] good performance status and minimum radiation dose of 60 Gy (but some did not get that) and at least the equivalent of ≥2 platinum doses. PET, who knows? In this country, these patients would routinely get PET, but we have no data on their original PET, so for all we know some [patients] extensively had stage III disease, but in reality may have had a cult metastasis.
Progression-free survival [PFS] and overall survival [OS] were the primary endpoints. You see a whopping PFS benefit, a 3-fold improvement in PFS5.6 to 17.2 months—and that holds up at 18 months. It’s about a 20% [survival] advantage, with a unprec­edented hazard ratio [HR] of 0.51.
The forest plot in the firstNew England Journal of Medicinepaper shows that pretty much every subgroup benefited.3 I think this is the only trial I’m aware of that had 2 separate New England Journal of Medicinepapers; they were presented by Luis Paz-Ares, MD, PhD, but published by Scott Antonia, MD, PhD, who was at Moffitt at the time. Regardless of histology, [there was a benefit, although] nonsquamous patients may have had a slightly higher benefit. In the original paper, PD-L1 expression of 25% was the cutoff point. Above that saw somewhat better benefita HR of 0.41, and below that was 0.59, but still a benefit. What’s miss­ing is a PD-L1 0% group. Forty or so patients, about 5% to 6% of those enrolled, hadEGFRmutations. They don’t seem to benefit as muchthe confidence intervals are overlapping unity, but [the numbers] still favor durvalumab. [Because the number of patients was so] small, the confidence intervals are going to be wide.
This has also changed the paradigm. Usually we wouldn’t repeat a scan until 6 or 8 weeks, but in this study, they got scans immediately after chemoradiation was done. If you’re going to abide by PACIFIC, you’re going to do a scan sooner than you would ordinarily have done in the past. I buy the fact that if [the patient’s] last radiation from randomization was less than 2 weeks, [the patient did] better. But that may be a self-fulfilling prophecy. These were the folks who probably had smaller-volume tumors, didn’t have as many complications, and were ripe for randomization. If they had complications, maybe a bigger tumor or something else going on, they did a little bit worse. It’s an interesting observation that speaks to the possibility that maybe the proximal use of checkpoint inhibitors immediately after chemoradiation might be better; that’s been argued. But there are a lot of unknown variables here.
The survival was updated at ASCO [the American Society of Clinical Oncology Annual Meeting] this year by Jhanelle Gray, MD, at Moffitt.4It’s holding up the survival advantage. The HR is 0.69, and at 3 years, [the OS rate was] 57% versus 43.5%so, again, unprecedented. I don’t know of any prior trial that had a median survival that reached 3 years, let alone exceeded 3 years.
The clock starts ticking after the chemoradiation, so effectively you’ve culled out those who wouldn’t do well during chemora­diation because of toxicity or disease progression, and I guess that’s about 20% to 30% of the population. So it’s not directly comparable with the old CALGB studies or RTOG or even 0617.
Targeted Oncology: What about some of the subgroups studied in the PACIFIC trial, including those with low PD-L1 expression? Did they also benefit from durvalumab?
Langer:No matter which [cancer] site [you looked at, patients] had a benefit: There were decreases in lung recurrences, extrathoracic recurrences, and brain recurrences. Remember, we invested a lot of time and energy in a big study looking at prophylactic cranial radiation, which was negative, but here you see a roughly 50% reduction of brain recurrence, so clearly immunotherapy is working broadly.
What about that PD-L1 <1% group? The second article in theNew England Journal of Medicineincludes it, but not in the main paper, in the supplements.5It’s a small group, about 150 or so of the 700 who were accrued, and the HR is on the wrong side. This is a post hoc exploratory analysis; fewer than 60% of patients in the study actually had adequate tissue for PD-L1 assessment. Those are relatively small numbers, and what’s intriguing is that at ASCO last year the HR had dropped from 1.36 to 1.14. There is a lot of fluidity here. I view this as hypothesis-generating. In the United States at least, it’s approved across the board, regard­less of PD-L1 status. In the European Union, it’s approved only for 1% or higher based on this analysis, even though [the analy­sis didn’t appear with] the main data.
Targeted Oncology: Does this finding affect how you treat patients in a similar setting?
Langer:I have to confess that I give it with some ambivalence. If the patient had a remote history of Crohn disease I probably would give it, whereas if they had 80% PD-L1 and a remote history of Crohn disease I would definitely give it. If they’re 1% or higher, I embrace it with enthusiasm. If they’re 1% or lower, I’ll have some ambivalence, but I’ll offer it to patients. The HR at this level was still favoring PFS at least, and it’s an exploratory analysis.
Targeted Oncology: How long would you be willing to wait after radiation to begin giving durvalumab? Is 6 weeks inviolate?
Langer:The study stipulated 6 weeks. Six weeks is not inviolate, but we try to do it within reason.
Targeted Oncology:Would you give durvalumab to a patient with anEGFRmutation?
Langer:I have given it, but I made it clear that we were in a data-free zone, so it was a mutual decision. I personally would consider it. The jury is completely out; it’s small numbers, and we haven’t seen the survival. The 1 argument against giving it, and this is more of a theoretical argument, is that if they relapse and they need osimertinib [Tagrisso], some recent data indicate that you increase the likelihood of toxicity if you give a checkpoint inhibitor to someone with an onco­genic driver and then follow that with targeted therapy. Why are we trying to anticipate recurrence when we’re trying to prevent recurrence?but if they do recur, then it becomes a bit vexing.
Targeted Oncology:What is the toxicity profile of durvalumab from the PACIFIC trial?
Langer:Pneumonitis was the big concern. It was higher overall at 34% versus 25%, but grade 3/4 pneumonitis rates were low, 3.4% versus 2.6%. My own personal experience is a bit higher, but a lot of the patients wouldn’t necessarily have gotten on this trial.
If you look at the any-grade adverse events, [the rates are] a little bit higher68% versus 53%—and of course, immune-mediated adverse events were 24% versus 8%. The vast majority were grade 1/2, easily handled, and pneumonia of any grade, whether it was pneumonitis or any other toxicity, was 13% versus 7.7%. It’s manage­able; it is increased, but it’s only marginally worse.
Targeted Oncology:How has this treatment been incorporated into standard clinical practice?
Langer:The new NCCN guidelines, as of the beginning of the year, include durvalumab after chemoradiation.1It’s given every 2 weeks for up to a year.