Research Efforts Mount With Immunotherapy in Advanced NSCLC

Supplements And Featured Publications, My Treatment Approach: Lung Cancer Practice Patterns, Volume 1, Issue 1

Christopher G. Azzoli, MD, discusses the importance of identifying molecular aberrations prior to starting a patient with advanced non–small cell lung cancer on immunotherapy, ongoing research aimed at furthering the immunotherapy paradigm, and explained why novel biomarkers of response to immunotherapy are needed.

The field of advanced non–small cell lung cancer (NSCLC) has grown significantly since the integration of immunotherapy, with pembrolizumab (Keytruda) remaining the mainstay frontline treatment for most patients who lack driver mutations, said Christopher G. Azzoli, MD, who added that although combining immunotherapy with targeted therapy has demonstrated significant toxicity, there may be clinical relevance in identifying how a patient’s genetics can inform responses to immunotherapy.

“What is lacking [with targeted therapy and immunotherapy] is the synthesis where genetic testing informs immunotherapy selection,” said Azzoli. “If we ever get that [data], we are off to the races for universal comprehensive testing. The biggest advancement in NSCLC was immunotherapy. If genetics can inform [responses to] immunotherapy, we will have locked in comprehensive testing, perhaps even reflexively.”

Additionally, refining immunotherapy in advanced NSCLC is the subject of significant clinical research, Azzoli explained. For example, ongoing research evaluating novel immunotherapies to improve immune responses, biomarkers of response beyond CT scans, and ways to overcome acquired resistance to targeted therapy and pembrolizumab are poised to further expand the paradigm.

In an interview with OncLive®, Azzoli, director of Thoracic Oncology at Lifespan Cancer Institute, discussed the importance of identifying molecular aberrations prior to starting a patient with advanced NSCLC on immunotherapy, ongoing research aimed at furthering the immunotherapy paradigm, and explained why novel biomarkers of response to immunotherapy are needed.

OncLive®: Why is it important to established whether a patient has an actionable driver alteration prior to initiating immunotherapy?

Azzoli: The biggest advance for the majority of our patients [with advanced NSCLC] was the discovery of PD-1 checkpoint inhibitors, which have doubled overall survival. [These patients make up] the majority of smoking-related lung cancer, and [these cancers] have many oncogene mutations and many neoantigens.

The minority [patient population comprises] never-smokers with lung cancer or lung cancer caused by an actionable mutation. The early efforts to combine checkpoint inhibitors with gene targeted therapy failed. When we give an EGFR or ALK TKI with a checkpoint inhibitor, we see toxicity that interferes with the delivery of the targeted therapy. That’s taking an excellent therapy and interfering it with an immunotherapy, which causes toxicity. Patients appear to have worse outcomes [with the combination of immunotherapy and targeted therapy] because [the immunotherapy is] interfering with the effective targeted therapy without a signal that the immunotherapy is helping. That is why we would be doing a patient a disservice [to give immunotherapy with targeted therapy].

If we give a patient first-line pembrolizumab and then discover they have an EGFR mutation, pembrolizumab has a half-life of 28 days. If you want to start the targeted therapy, because pembrolizumab is on board, [the patient will have] toxicity. [Immunotherapy] interferes with the safe delivery of targeted therapy.

One of the reasons for this distinction between immunotherapy and targeted therapy is [non-]smoking and smoking-related lung cancer. Everybody is wondering as we are looking forward to targeted therapy for smoking-related genetic [mutations], such as KRAS G12C, whether we will be able to mix immunotherapy and targeted therapy for patients who are more likely to benefit from immunotherapy because they have more smoking-related neoantigens.

This [field] is evolving, but as of today, pembrolizumab interferes with targeted therapy, which is why we need to get the diagnosis [of the patients’ disease] right out of the gate. We don’t want to interfere with a targeted agent.

What protocols are in place to ensure that patients receive the right therapy whether that be targeted therapy, immunotherapy, or chemotherapy?

There are 2 issues with molecular pathology: ruling in and ruling out. We have to rule out [the presence of] EGFR and ALK [mutations] so that we can give first-line immunotherapy. We have to rule [those mutations] out so that we can get the right drug to the right patient.

Let’s say that that we rule in [the presence of driver mutations] and get rapid results [that identify] a MET exon 14 skipping mutation, RET fusion, EGFR mutation, et cetera. No matter how fast we rule in, drug access [remains a] rate-limiting step in the initiation of therapy.

I have to congratulate my partners in industry because they realize [this barrier] and make sure that patients always get the drug. [This is also because of the efforts of] my access nurse and access team. My most valuable partner in my practice is my access nurse for oral chemotherapy because there is so much paperwork involved in getting these drugs quickly. They are high-cost drugs in which the initial prescription is [often accompanied by] an unaffordable copay. Some patients may not have insurance at all.

Thanks to our partners in industry, access seems to work well, but I’m grateful to have a nurse whose full-time job it is to handle access when I rule in for one of these druggable oncogenes.

Regarding ruling out, I always keep track in my head of how I’ve ruled the patient out [for a driver mutation]. Perhaps it was because their blood tests came back negative, or their tissue was insufficient. I’m always wondering whether the blood test was a false negative and when I’m going to rebiopsy the patient to become more confident that [I can] rule them out [for a driver mutation]. I keep track of my patients, including what [tests] I’ve sent, what results have come back, whether it was comprehensive, and what platform was used.

The task of keeping all of this on track rests on the medical oncologist. Our pathologists are trying to help, but they are not involved in the clinical context of the patient to understand when we might want to rebiopsy.

Should all patients with advanced lung cancer receive comprehensive genetic testing?

The number of actionable mutations and actionable genes has grown to the point of where we can no longer cherry pick molecular pathology. There are different methods, [one of which says] we can do EGFR and ALK [testing] in house and hold off on a comprehensive panel. However, we are getting to the point where we need rapid and routine comprehensive testing.

We can’t ignore the fact that we can’t [conduct] stepwise testing because if we [identify] an actionable mutation, such as a KRAS G12C mutation, [that] affects treatment decisions.

What remains unknown regarding acquired resistance to targeted therapy?

This is a golden age for clinical research in lung cancer because of all the different avenues for discovery and refinement of existing therapies. [Lung cancer] is one disease; it presents in similar fashion whether it is oncogene driven or smoking related. We have so many new drugs; there are [numerous] FDA-approved, commercially available drugs for lung cancer.

One conundrum is acquired resistance to targeted therapy. We can take that conversation to any [oncogenic] diagnosis, including EGFR, ALK, RET, MET, and so on, [to then look at] new drugs for acquired resistance to best initial targeted therapy. We are seeing a huge number of phase 1b/2 studies looking at novel immunotherapies after pembrolizumab, which remains the most popular first-line immunotherapy.

What key research is ongoing to further the paradigm for patients who lack actionable alterations?

I work in a smaller market in Providence, Rhode Island where we don’t have enough patients with specific gene targets to enroll in targeted-drug trials. [Therefore], I send most of my patients with a unique, druggable oncogene to larger cancer centers in Boston or New York. However, I see a lot of patients with smoking-related lung cancer without druggable oncogenes. Under my direction, we are launching mostly phase 2 trials of novel immunotherapies. We are [working with] industry partners to open up these studies that we know will accrue quickly because we have a lot of patients in need [of novel immunotherapies].

Going into [the 2021 ASCO Annual Meeting], we will start to see the results of those phase 2 trials and get a sense of which drugs work and [moreover], which drugs may work after pembrolizumab. Those early efficacy signals could drive enrollment to the current trials.

We also have ongoing phase 1 [trials] with so many different approaches to novel immunotherapy. [For example, trials are ongoing, evaluating] intratumoral therapy to improve immune responses to existing [therapies], auto-vaccination techniques vs combinations of checkpoint inhibitors, and combination [strategies] to stimulate immune responses and turn off regulatory responses that protect cancers through host defenses.

How can the field improve upon biomarkers of response to immunotherapy?

The drugs come from industry, so the drugs and protocols arrive at our doorstep and the studies need to be done. Investigators around the world pitch in to get these studies open and done, and our patients, graciously participate.

I’m unsatisfied with the CT scan as the biomarker of response. Too many patients are getting immunotherapy after radiation where we don’t know whether [the immunotherapy] is working. A lot of patients are getting immunotherapy combined with chemotherapy, where we also don’t know whether [the immunotherapy] is working. We don’t know whether the patient is having an immune response.

I have faith that there will be a diagnostic test—likely a blood test—to add to the CT scan [and provide insight into whether] a patient is having T-cell–mediated efficacy. That will be important also when patients get sick during immunotherapy so that we [can determine whether] it is an autoimmune process or an infection that is making them sick.


Because I am at the bedside with my patients, I see [the lack of biomarkers for response to immunotherapy] as a major unmet need. I’m diving into this research, collecting blood samples before and during immunotherapy, and trying to find a signal of T-cell–mediated anticancer activity.