Data From the PACIFIC Trial have Changed the Management of Stage III NSCLC, Observes West

H. Jack West, MD

H. Jack West, MD

H. Jack West, MD, recently discussed the treatment options and considerations he makes when treating patients with stage III and IV non—small cell lung cancer (NSCLC). West, medical director of the Thoracic Oncology Program, Swedish Cancer Institute, explained the nuances that go into his treatment decisions when discussing 2 case scenarios during aTargeted Oncologylive case-based peer perspective presentation.

Case 1

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, well-managed on simvastatin (Zocor); hypothyroidism, managed on levothyroxine (Synthroid); chronic obstructive pulmonary disease, managed on inhalers. He had a 40-pack-year smoking history, but recently quit. A physical exam was performed and showed intermittent wheezing, but he had an ECOG performance status of 1. His creatinine clearance levels were within normal limits.

A chest x-ray revealed opacity in the lung right upper lobe and a chest CT showed a 3.1-cm spiculated mass in the right upper lobe and 2 enlarged right mediastinal lymph nodes measuring 1.5 cm and 1.7 cm; moderate emphysema noted. A PET confirmed lung lesion and mediastinal lymphadenopathy without evidence of distant metastases. His brain MRI was negative.

A bronchoscopy with transbronchial lung biopsy and lymph node sampling revealed adenocarcinoma with positive nodes in stations 4R and 7; level 4L was negative.Genetic testing was negative for known driver mutations. He was staged with T2aN2M0, stage IIIa.

Targeted Oncology: What are your general impressions of this case?

West: This patient was staged at T2aN2M0, stage IIIa. That is arguably in the range where you could consider a few different approach options, both surgical and nonsurgical. Based on his mediastinal disease, PFT, emphysema, and the overall look of the patient, he was felt to be not as well served by surgery and was referred for concurrent chemotherapy and radiation. This is [what happens in] the majority of [patients with] stage III disease; operable stage III cases are in the minority.

Do you typically order genetic testing for locally advanced lung cancer?

There is no right or wrong answer. I would say that the board answer is there is no standard role for molecular testing at this point. Targeted therapies, and specifically the immunotherapies that we are going to talk about, are not gated by PD-L1 status. That is not to say that people will not order it at times, but it is not clear what you can do with it.

What are the options for treatment for this patient?

As I said before, there is the concept of resectable stage IIIa and II disease with mediastinal node involvement. Some experts in the field feel that this term of “operable” is almost meaningless, because if you have a surgeon who is aggressive enough, everything is operable. That doesn’t mean it’s a good idea.

There are 2 options—neoadjuvant chemotherapy or chemoradiation—and there is no “best” one. Some trials use chemotherapy alone, while others use chemoradiation. I was the primary investigator of an ill-fated stage III trial, which tried to compare chemotherapy with chemoradiation as induction. It closed with more investigators than subjects. There are some questions that are hard to answer.

In contrast, there is a clear answer to the question of chemoradiation concurrently or sequentially. For patients who are fit enough, the survival with concurrent chemoradiation instead of sequential has been shown to be better in North American as well as Japanese trials. That is a pretty established standard.

The real question has been, “What can we do to improve upon the old standard of giving 7 weeks of chemotherapy and concurrent radiation?” There has been this consensus that 7 weeks can’t be enough if we are routinely giving 4 cycles or 3 months of chemotherapy in the adjuvant setting for a stage I or II cancer. Yet we haven’t seen any proven value to giving additional chemotherapy before or after the concurrent portion. That generally adds toxicity. We’ve tried to give agents like gefitinib (Iressa), and that led to detrimental effects; significantly better results were seen with placebo in unselected patients. We also tried increasing the radiation, and that also led to worse outcomes by escalating the dose. The question is, can we ever get beyond the impasse of 20% cure rates with chemotherapy and radiation for unresectable disease?

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 radiotherapy. 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.

What are your general impressions after the treatment of this patient?

This patient underwent a typical approach of cisplatin/etoposide, along with concurrent thoracic radiation. That would typically be 61 to 66 Gray (Gy); 74 Gy [has been shown to be] too toxic and led to worse outcomes [in trials].

Follow-up imaging showed a partial response with shrinkage of the primary and nodal lesions. One important thing I tell my patients before they get their scan is that you can’t tell what’s going on initially after the chemotherapy and radiation. It’s a murky mess because the radiation, along with the chemotherapy, causes inflammation. You don’t know what is dead, dying, or still viable—it is initially very ambiguous.

Does this patient require further treatment?

The big news is the PACIFIC trial, which was the leading presentation at the European Society for Medical Oncology (ESMO) meeting in September 2017. It had a concurrent publication in theNew England Journal of Medicineby Dr Scott Antonio and colleagues.¹It broke the impasse of more than a decade of no real improvements in how we manage patients with stage III disease.

The study design was for patients with locally advanced, unresectable NSCLC. It was a 2:1 randomization to consolidation durvalumab (Imfinzi) every 2 weeks for up to a year or placebo, on that same schedule. We have seen progression-free survival (PFS) presented; however, we have not yet seen the overall survival (OS) data.

The PFS was highly significant with a hazard ratio of 0.52. Unlike so many positive trials where we might say, “Statistically positive, but clinically not so much,” this was a huge difference. There was a 3-fold difference in the median PFS, just 5.6 months with placebo versus 16.8 months. Additionally, there was a 17% difference in PFS at 18 months. We have learned, just in the past few months, from a press release, that this trial is also positive for an OS benefit, but we haven’t yet seen those data.

There was a range of chemotherapy options that were allowed, including platinum-based doublet for at least 2 cycles, but you could not get postradiation chemotherapy. Patients potentially received induction chemotherapy before the chemotherapy and radiation. That was permitted, and about 22% of the patients got that.

If the patients on the trial had progressed, they would have gone straight to durvalumab or placebo. [Patients did not receive] anything in between treatments with the idea that the radiation may be a sensitizer to the immunotherapy.

Did any subgroups benefit more than others in the PACIFIC trial?

There are various additional subgroups and factors that you may wonder, is there a difference [in benefit for these populations]? Histology didn’t make a difference. Both the squamous and nonsquamous patients had benefit. Regarding PD-L1 status, in the setting of stage IV disease, we have frequently seen that the high PD-L1 patients are the ones who get the most benefit from the immunotherapy. But, in other trials, previously treated patients do benefit sometimes from immunotherapy, and certainly in combination with chemotherapy.

In this trial, those who had low or high PD-L1 at the 25% cutoff got the same benefit. No matter what the [status], durvalumab is appropriate. If it is unknown, high, zero—it doesn’t matter. There is no clear indication for doing PD-L1 testing in this setting.

Some people argue aboutEGFRmutation status. In general, patients with anEGFRmutation in stage IV disease have tended to be the patients who get the least benefit [from immunotherapy], but that is not to saynobenefit. In some combination trials that are coming out, the patients with anEGFRmutation get a benefit that is very comparable with or even better than that of the rest of the population.

The mutation-positive patient subgroup crosses 1, but this is still a small subset. The hazard ratio is 0.76, and I would say that this is not what these subset analyses are designed to do. They are not powered to show statistical significance in every subgroup. Still, there are clinicians who would still say, “I would withhold this therapy for anEGFRmutation—positive patient.” I would say that this is not what the data tell us. These patients were included in the trial and the trial was very positive. Even though they may have had a less positive result [than other patients, the trial was] too small to say anything conclusive.

What was the significance of the short time period between chemoradiation and immunotherapy in the PACIFIC trial?

There has been some discussion about the finding that the patients who had a shorter interval between stopping radiation and starting durvalumab did especially well. But this was an evolution of the trial. Initially, they had a very narrow window of 14 days and they had trouble getting patients enrolled. They had to broaden it to 6 weeks. But I would say that there may be something different about those who complete chemotherapy radiation, which is challenging, and then are ready within the next 2 weeks to go on to their next treatment. They were often more fit, they may have smaller radiation fields, etc. They were not randomized to [earlier treatment with durvalumab] and it could be a selection bias.

I don’t know if the timing here is optimal or not. I am not convinced it is that important to do it within 2 weeks. In fact, my standard has been to get scans about 3 to 4 weeks after chemoradiation because the radiation is still shrinking [the tumor] for a while afterward. In fact, the response rate from the trial just came out, and it showed a 16% response rate on placebo. To me, that just shows that there is ongoing shrinkage after a very early scan. I have begun doing my scans earlier, 2 to 3 weeks instead of 3 to 4 weeks. I feel no compulsion to get a scan 8 days after they do chemoradiation so that I can have them start [immunotherapy] on day 10. I don’t think that it matters at all. Probably, the patients who were able to rush right on to [durvalumab on the trial] are different from most patients who would like a few days to recover from the chemotherapy and radiation. They are kind of crawling to the finish, still having trouble swallowing, and I think it is perfectly OK to wait an extra 3 to 4 weeks and start [immunotherapy] 4 or 5 weeks after chemoradiation.

What were the toxicities associated with treatment on the PACIFIC trial?

Certainly,with immunotherapy as a chaser after chemoradiation, we could be concerned about that, but reassuringly, there were few overall differences in toxicities. There were [somewhat more frequent occurrences of] immune-related adverse events with durvalumab. The biggest issue was pneumonitis, but importantly, pneumonitis can occur even when you don’t receive durvalumab. There were almost no differences in toxicities between the placebo arm and durvalumab arm in patients with grade 3 or 4 NSCLC. It was reassuring that you didn’t get frequent pneumonitis in these patients. It occurred a little more frequently [in the durvalumab arm], but we have been dealing with pneumonitis in these patients in the 2-, 3-, 4-month-long period after they complete radiation and chemotherapy for a long time—long before durvalumab came on the scene.

While patients who [began immunotherapy] within 14 days [of completing chemoradiation] seemed less likely to develop pneumonitis, we don’t know that it is a causal effect of the timing of durvalumab. It may just be an indirect issue, in which the patients with smallest radiation fields are the best able to go straight to immunotherapy. Definitely, there is a high association between the volume of radiation field and the risk of pneumonitis. There are a lot of confounding factors here.

What is the significance of the PACIFIC trial?

We will get OS data within the next few months. But based on [the initial] results alone, the FDA approved durvalumab about 4 months ago. Nearly all of us were inclined to get durvalumab for our patients in this setting as we soon as we saw the data. Now, it is the standard of care for patients who are well enough to do it.

The main questions are: Should we think about starting immunotherapy earlier with the chemotherapy or as induction? Should we be using immunotherapy sequentially for our patients? But for a very common scenario of unresectable stage III chemoradiation, this is the post treatment, and the [FDA] approval is for a year of treatment.

Do any other trials show promise?

The Hoosier Cancer Research Network trial based out of Indiana, done in the last couple of years, was basically the exact same kind of study, but with pembrolizumab (Keytruda) every 3 weeks.²They saw the same kind of results overall—nothing different, nothing better. One advantage is that the drug was administered every 3 weeks, but none of these results in absolute terms were better than what we have seen. It was also a trial of fewer than 100 patients. It is very possible that other drugs will be tested in this setting.

Case 2

An 81-year-old man presented with symptoms of coughing, dyspnea, upper back pain, and fatigue requiring frequent rest. His past medical history revealed hypercholesterolemia, controlled on pravastatin (Pravachol); hypertension, controlled on verapamil (Verelan); psoriatic arthritis, not on treatment for 3-plus years. He was a former smoker, but physically active and played golf most weeks with an ECOG performance status of 1.

A Chest CT revealed a 2.5-cm solid mass in the left upper lobe and lymphadenopathy in the left hilar and bilateral mediastinal nodes and bilateral, small pulmonary nodules, with the largest one 8 mm. PET/CT imaging showed 18F-FDG uptake in the lung mass, left hilar and both mediastinal lymph nodes, and thoracic spine (T5/T6).

A bronchoscopy and transbronchial lung biopsy were performed and pathology showed grade III squamous cell carcinoma; PD-L1 expression by immunohistochemistry 22C3, tumor proportion score of 65%. He was later diagnosis stage IV squamous NSCLC.

What are your general impressions of this patient?

Here’s a case of metastatic squamous NSCLC. For board review purposes, squamous accounts for 20% to 25% of the lung cancer patient population. It is second most common. I would say that these patients have a disproportionately significant smoking history and many comorbidities. He has a high PD-L1 expression, which is seen in about 30% of NSCLC cases.

Are you routinely testing for PD-L1 expression in newly diagnosed patients?

Yes, although I will say that we will be seeing some data that show that there are certain treatment options that you can do that are available and approved for everything on the spectrum. People may ask, “Do we need to do this anymore?” But I would say that there is still good reason.

The patient was started on pembrolizumab.

Is there a potential role for radiation therapy to address his symptoms of dyspnea or back pain?

Radiation to the spinal lesions could make sense in this case for palliative reasons, if it is bad enough. Without images to look at, I am not sure that dyspnea is likely to be amendable unless there were a lot of lymph nodes that seemed to be causing obstructive symptoms.

What are the options for systemic therapy?

At this point, the leading options are monotherapy with pembrolizumab or chemotherapy/pembrolizumab. This patient was started on monotherapy pembrolizumab.

How do his older age and good performance status factor into this choice?

I would say that those factors don’t make that much of a difference. I would not be incredibly eager to give concurrent chemotherapy and immunotherapy to an 81-year-old if you don’t have to, but he would have been eligible for these [kinds of] trials.

These trials were for patients with a performance status of 0 or 1. I would be hesitant to give chemotherapy and immunotherapy triplets to [patients with] a performance status of 2 or 3, but immunotherapy we have generally translated for [patients with a] performance status of 2 at least. We presume that is just as good or nearly as good, but we don’t have data.

Do precautions need to be taken considering his psoriatic arthritis?

We don’t know, but I would be following this issue. About 10% to 15% of the patients in our clinic have some autoimmune disorder, serious or not. We have to weigh that in the decision. [For example,] I have a patient with ulcerative colitis with anEGFRmutation, so I have been hesitant to give her immunotherapy. She is not that likely to benefit, and I don’t want to see how bad that can get if she is requiring active management for ulcerative colitis.

What is the issue surrounding the use of steroids?

There was a presentation at the 2018 American Society of Clinical Oncology (ASCO) Annual Meeting by Memorial Sloan Kettering.³They looked at the outcomes of patients who got prednisone at 10 mg daily chronic or higher equivalent, and also got immunotherapy. These patients were compared with those who weren’t on steroids.

The results showed that all endpoints were worse in the patients on chronic steroids, leading us to some concern that you don’t want to give steroids gratuitously, although very few people are on them. The caveat is that these patients were on steroids for chronic fatigue, pain, and brain metastases, and those issues in themselves are associated with less favorable outcomes. We don’t know completely how much of this is from the steroids and how much of this is from the reason they needed the steroids.

In general, I would recommend against giving steroids just to cover the bases. But the belief is that if someone needs steroids for controlling adverse events (AEs), don’t hesitate. As we’re seeing in subsequent work, regimens that require steroid premedication for a couple of days around the chemotherapy have been perfectly effective. This chronic steroid issue does not apply to taking 3 days of steroids with pemetrexed.

What is the rationale for the use of pembrolizumab in this patient setting?

KEYNOTE-024 was presented at the European Society for Medical Oncology Annual Meeting in 2016 and published in theNew England Journal of Medicine.⁴Based on this trial, the FDA took only 3 weeks to approve pembrolizumab monotherapy. This was just for patients with stage IV NSCLC and PD-L1 of 50% or greater. About 30% of the patients were squamous or nonsquamous, and patients could not have anEGFRmutation or anALKrearrangement. They were randomized 1:1 between histology-appropriate chemotherapy or pembrolizumab monotherapy. It showed a very clear benefit for pembrolizumab.

A few points to make from this: First, this is clearly an effective therapy that can work for years for patients. It is the minority of patients who [experience] immune-related AEs. One unsettling characteristic is that we have had decades of experience dealing with chemotherapy AEs, and these AEs are more idiosyncratic. Yet, at this point, [pembrolizumab] is the standard of care.

Are there data to support the use of pembrolizumab for other patient populations?

This was the background for a plenary session at ASCO this year, called KEYNOTE-042.⁵KEYNOTE-024 was for the patients with 50% or higher PD-L1—that is, about a third—and this trial opens it up to another third who have 1% or higher PD-L1. The last third are PD-L1 less than 1%. This is now looking at a broader population with the exact same question of pembrolizumab monotherapy versus histology-appropriate doublet chemotherapy and cutting out the patients withEFGRmutations orALKrearrangements because those patients have other therapies and don’t seem to benefit much from this.

The study design is a 1:1 randomization between pembrolizumab and either carboplatin/paclitaxel for the squamous or carboplatin/pemetrexed for the nonsquamous. The breakdown of the patients on the trial is not an even spectrum: You have half for the greater-than-50%, and the other half are mostly concentrated under 20%. You don’t get a lot of 48%, because if the pathologist saw 48% they would call it 50%. So, you don’t get a lot [of patients in] the 20% to 50% range.

These trials keep the best patients. Although the topline results and ASCO headlines say pembrolizumab is a great option for anyone with PD-L1, that’s not the right answer. The real answer is this group of PD-L1 [expression level] is so benefited by pembrolizumab that this trial is positive despite adding in patients who don’t benefit.

Are there any other caveats to remember with this trial?

Another important aspect is that there was no crossover allowed. It has been a standard of care for 3 or 4 years that patients who start on chemotherapy would get a second-line checkpoint inhibitor. There are 3 of those approved: pembrolizumab, nivolumab (Opdivo), and atezolizumab (Tecentriq). They all have a clearly proven survival benefit and are standard of care, and yet [patients] were not allowed to get that in the trial after progression on the chemotherapy arm. That is a real handicap.

While some clinicians came out of this saying, “Patients did just as well, so I feel like I can still give pembrolizumab,” [what happened is] that [the patients did] just as well with the chemotherapy arm, [but the chemotherapy arm was] severely handicapped on this trial and patients were not truly getting the standard of care. Only 20% of patients in this global trial ever got immunotherapy after chemotherapy. That is not a fair fight. And it is disappointing for first-line pembrolizumab in that setting. The other thing that needs to be considered when we’re talking about KEYNOTE-042 are the other options out there. 

Could combination pembrolizumab/chemotherapy be considered for this patient?

KEYNOTE-407 was a phase III trial in squamous NSCLC with any PD-L1 level allowed.⁶It randomized patients to carboplatin with either paclitaxel or nab-paclitaxel (Abraxane) every 3 weeks and placebo/pembrolizumab. In the pembrolizumab arm, you did get maintenance with pembrolizumab. There was crossover allowed in this trial.

This was positive for an OS benefit favoring the chemotherapy/pembrolizumab arm. What we had heard in the abstract is that the response rate was better. [Ultimately], this trial did not disappoint. [The OS benefit] holds true for all subgroups, particularly with the paclitaxel or nab-paclitaxel. Both get the same benefit there. What I think is most impressive is that it holds up for PD-L1 across the board.

Have any other combinations been studied for this patient population?

IMpower131 was a study of carboplatin and nab-paclitaxel for all patients with or without atezolizumab.⁷It is a large trial with more than 1000 squamous patients and any PD-L1 level. The control arm was carboplatin and nab-paclitaxel for up to 6 cycles followed by supportive care, and the same regimen followed by atezolizumab. The third arm was carboplatin/paclitaxel with atezolizumab. That was the hierarchical, statistical design that they never showed the results from, because that would only be unlocked if the trial was positive for OS.

Co-primary endpoints were PFS and OS. The PFS is statistically significant and holds up against a lot of different subgroups. Across the PD-L1 spectrum, they saw what they often see in these trials. The high PD-L1 patients got the biggest benefit.

Additionally, the response rates are higher, but modestly so. The disappointing thing is that it did not show a difference in OS. The hazard ratio is 0.96. There is some separation of the curves by the time that you get up to 2 years, but for most of that curve they are traveling neck and neck.

Do you offer combination therapy to your patients?

We are in a situation where we don’t have an FDA-approved option of chemotherapy/pembrolizumab. But to me, the best choice is the KEYNOTE-407 regimen of carboplatin with the taxane of your choice and pembrolizumab. I would say that that is the best choice for at least PD-L1 less than 50%. If it is 50% or higher, my general approach would be to favor the single agent unless patients have a high tumor volume or rapidly progressing disease.

Otherwise, if it is a patient who is fit, I am going to favor immunotherapy alone as the first choice for patients with PD-L1 high expression and chemotherapy/immunotherapy for those with PD-L1 low expression. The KEYNOTE-407 trial is going to be the best treatment choice right now. Everything else, such as the trial of IMpower131, is positive, but the comparative today is not just chemotherapy doublet: Instead, it’s “Is it better than what we just saw?”—and it isn’t. So, I would say the new king of the hill is carboplatin/paclitaxel, or nab-paclitaxel and pembrolizumab for the squamous.

Do you believe that there is any role for first-line chemotherapy alone?

Not much, these days—just for the patients who are not good candidates for immunotherapy because they have significant comorbidities. I saw a patient who is 90 years old with a PD-L1 of 30%, and I would rather give her a significantly attenuated dose of carboplatin with full-dose pembrolizumab and cross our fingers than give her chemotherapy alone. Because if you get a hit on immunotherapy, you can sail on that for a long time.

References:

1. Antonia SJ, Villegas A, Daniel D, et al; PACIFIC Investigators. Durvalumab after chemoradiotherapy in stage III non-small-cell lung cancer.N Engl J Med. 2017;377(20):1919-1929. doi: 10.1056/NEJMoa1709937.
2. Durm GA, Althouse SK, Sadiq AA, et al. Phase II trial of concurrent chemoradiation with consolidation pembrolizumab in patients with unresectable stage III non-small cell lung cancer: Hoosier Cancer Research Network LUN 14-179.J Clin Oncol. 2018;36(15 suppl; abstr 8500).
3. Arbour KC, Mezquita L, Long N, et al. Deleterious effect of baseline steroids on efficacy of PD-(L)1 blockade in patients with NSCLC.J Clin Oncol. 2018;36(suppl; abstr 9003).
4. Reck M, Rodriguez-Abreu D, Robinson AG, et al; KEYNOTE-024 Investigators. Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer.N Engl J Med. 2016;375(19):1823-1833. doi: 10.1056/NEJMoa1606774.
5. Lopes G, Wu Y-L, Kudaba I, et al. Pembrolizumab (pembro) versus platinum-based chemotherapy (chemo) as first-line therapy for advanced/metastatic NSCLC with a PD-L1 tumor proportion score (TPS)>1%: open-label, phase 3 KEYNOTE-042 study.J Clin Oncol. 2018;36(suppl; abstr LBA4).
6. Paz-Ares LG, Luft A, Tafreshi A, et al. Phase 3 study of carboplatin-paclitaxel/nab-paclitaxel (chemo) with or without pembrolizumab (pembro) for patients (pts) with metastatic squamous non-small cell lung cancer (NSCLC).J Clin Oncol. 2018;36(suppl; abstr 105).
7. Jotte RM, Cappuzzo F, Vynnychenko I, et al. IMpower 131: primary PFS and safety analysis of a randomized phase III study of atezolizumab + carboplatin + paclitaxel or nab-paclitaxel vs carboplatin + nab-paclitaxel as 1L therapy in advanced squamous NSCLC.J Clin Oncol. 2018;36(suppl; abstr LBA9000).