CME
Physicians: Maximum of 1.25 AMA PRA Category 1 Credits™
Released: September 25, 2020
Expiration: September 24, 2021
Zofia Piotrowska, MD, MHS
Turning our attention to the targeted therapy space for advanced NSCLC, we will next discuss the ongoing, multicenter, open-label, parallel-group, randomized, phase III SINDAS study that was conducted in China.
SINDAS evaluated the addition of stereotactic body radiotherapy (SBRT) to an EGFR TKI in patients with newly diagnosed, advanced, EGFR‑mutated NSCLC with oligometastatic disease (up to 5 metastases, with none in the brain and no more than 2 lesions per organ).19 Patients (N = 133) were randomized to receive an EGFR TKI (gefitinib, erlotinib, or icotinib) with or without SBRT 25-40 Gy in 5 fractions. The primary endpoint was PFS, and the secondary endpoints were OS and safety.
This study is based on the observation that although patients with newly diagnosed EGFR-positive, advanced NSCLC do well with first‑line EGFR inhibitor therapy, acquired resistance will eventually develop, with the question being whether we can improve outcomes in the first-line setting by giving local therapy along with EGFR TKI therapy (or other targeted therapies). Previous studies have demonstrated that combining first‑line therapies of various types with radiation can be of benefit.20
The SINDAS study population was generally well balanced. However, I will point out that there was a higher rate of patients with EGFR L858R mutations (30% vs 16%) and a lower rate with exon 19 deletions (66% vs 72%) in the SBRT arm compared with the control arm. This is significant because, in general, patients with EGFR L858R tend to have slightly worse outcomes than patients with exon 19 deletions, and this could lead to an imbalance between the arms. The study also included patients with EGFR exon 20 insertions (4% in the SBRT arm and 12% in the control arm), which could also lead to an imbalance between the arms because patients with this mutation generally don’t respond well to first-generation and second-generation EGFR inhibitors.21
Zofia Piotrowska, MD, MHS
There was significant improvement in PFS (the primary endpoint) with the addition of SBRT to first-generation EGFR TKI therapy (median PFS: 20.2 months vs 12.5 months for the control arm; HR: 0.618; 95% CI: 0.394-0.969; P < .001). The secondary endpoint of OS was also improved from 17.4 months in the control arm to 25.5 with SBRT (HR: 0.682; 95% CI: 0.456-1.001; P < .001). Of note, the median follow-up of this study remains relatively short at 19.6 months, but we still see a survival improvement with adding radiation to EGFR TKI therapy.
Zofia Piotrowska, MD, MHS
Generally, the combination of an EGFR TKI and SBRT was well tolerated. There were no significant safety differences between the 2 arms, and no grade 5 events. Rash was the most common grade 3 AE but was balanced between the arms (n = 10 with SBRT vs 8 with control). Rates of other toxicities, including pneumonitis (n = 6 with SBRT vs 2 with control), were overall low.
Zofia Piotrowska, MD, MHS
Adding local therapy to EGFR TKI therapy, particularly for patients with oligometastatic disease or a low disease burden, is an intriguing approach and certainly has merit. However, when considering how to apply the SINDAS data to clinical practice, we must consider that this study used first-generation EGFR TKIs, which are the current standard of care in China, whereas in the United States the standard of care is to use third-generation EGFR TKIs (ie, osimertinib). Because SINDAS used a different EGFR TKI than what I currently use in my practice, I won’t necessarily use this approach for all of my patients going forward, but for select patients I am adding radiation therapy either at the time of diagnosis for oligometastatic disease, or perhaps more commonly, as consolidative therapy for patients who show a good response to an EGFR TKI and seem to be in their nadir. Ideally, I like to do this in the context of a clinical trial, but if one is not available I think it is worth considering in collaboration with radiation oncology colleagues.
Dr. Liu, what are your thoughts on this strategy?
Stephen V. Liu, MD
I agree that the SINDAS results are intriguing, and the data are consistent with findings from other studies. For example, data from 2 phase III randomized trials have shown that adding cytotoxic chemotherapy to first-generation EGFR TKI therapy in the frontline setting can improve survival.22,23 The working hypothesis behind this paradigm is that we can target clones that will ultimately become resistant to TKIs by adding a non–cross-tolerant therapy, like chemotherapy or radiation therapy, early in the therapeutic process, and that this aggressive initial therapy will eliminate many of those resistant clones upfront, thereby allowing the TKI to maintain disease suppression in the long run. I also wonder if it would be effective to combine chemotherapy plus a TKI followed by radiation to residual sites.
Zofia Piotrowska, MD, MHS
I am equally intrigued, but see the challenge now being how to incorporate this paradigm into clinical practice. Which strategy—adding chemotherapy or radiation or both—will really improve long-term outcomes? I talk to many patients about how we hope to turn EGFR‑mutated lung cancer into more of a chronic disease, where they are able to stay on each therapy for a long time, and I think the addition of radiation to a TKI is one way toward realizing that goal.
Zofia Piotrowska, MD, MHS
Staying on the topic of targeted therapy but moving to a different oncogenic driver, we will now discuss the updated registration dataset on pralsetinib in RET fusion–positive NSCLC from the ongoing, international, open label, phase I/II ARROW trial evaluating the selective RET inhibitor pralsetinib (BLU‑667) in RET-altered thyroid cancer, NSCLC, or other advanced solid tumors (planned N = 527).24 The phase I dose-escalation portion of ARROW has now been completed, establishing the recommended phase II pralsetinib dose of 400 mg/day, and phase II is ongoing. The primary endpoints are ORR and safety, and key secondary endpoints include duration of response, clinical benefit rate, and disease control rate.
The safety population in this analysis included 354 patients with any tumor type who were treated with pralsetinib at the recommended phase II dose. Of these, 179 had RET fusion–positive NSCLC and 175 had RET-altered solid tumors. The efficacy population for RET fusion–positive NSCLC comprised 132 treated patients in the ITT population and 116 in the response-evaluable population.
Zofia Piotrowska, MD, MHS
Consistent with previously reported data from ARROW,25 pralsetinib demonstrated impressive ORRs of 58% in the ITT population and 65% in the response‑evaluable population in this analysis. As we have seen with some other targeted therapies, the response rate seems to be higher in treatment‑naive patients: 66% in the ITT population (n = 29) and 73% in the response-evaluable population (n = 26). Response rates were lower but still compelling and impressive for the patients who received previous platinum-based chemotherapy (n = 92) at 55% and 61%, respectively.
Zofia Piotrowska, MD, MHS
The median duration of response was not reached (95% CI: 11.3 to not reached), but many patients have remained on treatment for 12 months or longer, with 75% of responders continuing treatment.
Zofia Piotrowska, MD, MHS
This waterfall plot shows very impressive rates of tumor shrinkage with pralsetinib, again regardless of prior treatment. In the overall response-evaluable population (n = 116), 96% experienced tumor reduction. Impressively, 100% of the 26 patients with no previous therapy demonstrated tumor reduction.
Zofia Piotrowska, MD, MHS
One of the key distinguishing factors of the selective RET inhibitors has been their CNS activity, which is very important clinically because many patients with advanced NSCLC present with brain metastases. As shown here, the intracranial ORR for the 9 patients who presented with measurable CNS metastases at baseline was 56%, with 3 patients achieving a CR. Although this is a very small dataset, these results suggest good CNS efficacy for pralsetinib, and most patients continue to be treated on the study.
Zofia Piotrowska, MD, MHS
Regarding the coprimary endpoint of safety, again, these data are consistent with previous reports for this drug.25 Generally, pralsetinib is well tolerated. The most common treatment-related AEs of any grade include aspartate aminotransferase and alanine aminotransferase elevations, anemia, constipation, hypertension, neutropenia, and some diarrhea. Of note, rates of grade ≥ 3 AEs were low, with only 4% of patients in the safety population discontinuing treatment. Of note, grade 3 or worse hypertension and neutropenia was present in 10% of patients, so those are certainly AEs to watch out for in these patients.
Zofia Piotrowska, MD, MHS
These data confirm that pralsetinib is a very effective therapy for RET fusion–positive NSCLC. In fact, on September 4, 2020, pralsetinib was granted accelerated approval by the FDA for the treatment of adult patients with metastatic RET fusion–positive NSCLC as detected by an FDA approved test.53 These results and approval are coming at a time when the FDA also recently approved selpercatinib, another highly selective RET TKI, for the treatment of patients with RET‑positive NSCLC.26 Both drugs seem to have high response rates, excellent CNS activity, and durable responses—it is very exciting to have new and effective therapies for our patients with RET-positive disease.
The addition of these new agents into the therapeutic landscape for advanced NSCLC underscores the importance of identifying the patients with NSCLC most likely to benefit from treatment with these agents. Broad molecular testing by next-generation sequencing, including for RET fusions, should be done at the time of diagnosis for every patient, especially considering the slightly higher response rates in the first-line setting with these RET inhibitors compared with later lines.
Dr. Lui, what is your take on these updated ARROW results?
Stephen V. Liu, MD
The ARROW data are very compelling. To me, this is exactly what targeted therapy should look like: responses in the vast majority of patients, with tumor reductions in 100% of those who are treatment naive, an impressive duration of response, and very well tolerated. Your point about needing to identify these patients so they are treated properly, is very well taken, especially now that emerging data suggest immunotherapy, which would otherwise be our standard, could be less effective for patients with driver oncogenes. Without proper identification of the RET fusion, patients with RET fusion–positive NSCLC are likely to receive immunotherapy, which is fairly ineffective in this setting, particularly as monotherapy.
Zofia Piotrowska, MD, MHS
As the list of oncogenic drivers with effective targeted therapies grows with each ASCO annual meeting, doing broad testing upfront will really help us to identify the patients who will benefit from each of these new and exciting therapies. In fact, I would say that maybe the most important take-home message from ASCO highlights that testing is critical for patients.
Zofia Piotrowska, MD, MHS
GEOMETRY mono-1 is an ongoing phase II multicohort study of the oral cMET inhibitor capmatinib in patients with stage IIIb/IV, MET-altered NSCLC and wild-type EGFR and ALK (planned N = 373).27 All patients received capmatinib 400 mg twice daily until disease progression. The primary endpoint was ORR, and the key secondary endpoint was duration of response, both assessed by blinded independent central review. Other secondary endpoints included investigator-assessed ORR and response duration, disease-control rate, PFS, OS, and safety.
At ASCO 2019, Wolf and colleagues28 presented an analysis of capmatinib in the cohort of patients with NSCLC and MET exon 14 skipping mutations. Capmatinib demonstrated clinically meaningful efficacy in these patients, leading to the accelerated approval of capmatinib by the FDA in June 2020 for patients with NSCLC and MET exon 14 skipping mutations regardless of the line of therapy.29
At ASCO 2020, Wolf and colleagues30 presented an analysis of capmatinib in a different cohort of patients, those with NSCLC and high-level MET amplification, which was defined as a gene copy number of at least 10. Results were reported for both pretreated cohort 1a (n = 69) and treatment-naive cohort 5a (n = 15).
Zofia Piotrowska, MD, MHS
For patients with NSCLC and high MET amplification, the ORR by blinded independent central review was 29% in the pretreated cohort and 40% in the treatment-naive cohort. Of interest, the median duration of response was 8.3 months in the pretreated cohort and 7.5 months in the treatment‑naive cohort, with the caveat that this group was comprised of only 15 patients.
Although these data show an important clinical benefit for capmatinib in NSCLC with high MET amplification, the rates are lower than what was demonstrated in the patient population with MET exon 14–skipping mutations28 and are lower than what is typically seen in patients with oncogene‑driven cancers who are receiving appropriate targeted therapy (eg, as discussed above for pralsetinib in RET fusion–positive NSCLC).24
Zofia Piotrowska, MD, MHS
The median PFS was also somewhat shorter than that seen with other targeted therapies at 4.07 months and 4.17 months in the pretreated group and treatment-naive group, respectively.30 The median OS in both populations was also similar, at approximately 10 months.
Zofia Piotrowska, MD, MHS
Safety was evaluated in the full dataset for MET-altered disease (N = 364) and was found to be consistent with previous reports for capmatinib. Most treatment-related AEs were grade 1/2, but serious treatment-related AEs occurred in 13.2% of patients. A total of 22.8% reduced their doses due to AEs, and 10.7% discontinued therapy altogether as a result.
The most common toxicities were peripheral edema (any grade: 42.9%; grade 3/4: 8.2%), nausea (any grade: 34.3%; grade 3/4: 1.6%), and vomiting (any grade: 18.7%; grade 3/4: 1.9%), which are important AEs to watch out for in patients treated with capmatinib. Peripheral edema, which is a class effect, can be particularly troublesome to patients and may require dose reduction or discontinuation.
Zofia Piotrowska, MD, MHS
Although MET amplification is not quite as important of a clinical biomarker as MET exon 14 skipping mutations or other oncogenes seen in clinical practice, I think it is relevant to test for, and it is included in many next-generation sequencing panels. If you do find a patient with MET-amplified NSCLC, it is important to look for whether they also have a MET exon 14 skipping mutation because these 2 genetic aberrations can overlap. So identification of MET amplification can potentially be useful in identifying a MET exon 14 mutation for which capmatinib is approved, as mentioned above. If they are truly just MET amplified, it would be reasonable to consider off-label use of capmatinib based on these data.
Stephen V. Liu, MD
I agree that MET amplification is an important biomarker, but nowhere near as important as the other identified drivers. Part of that is due to how MET amplification is defined and differences in where thresholds for high amplification are set; it requires a quantitative assessment vs the simpler qualitative assessment we use for other biomarkers. So while these data are welcome and the outcomes are good but modest, with a response rate of 30% to 40% and fairly limited PFS, I think it currently is more important to identify MET exon 14 skipping mutations in our patients. I am also interested to see if these data would apply to acquired resistance, where MET amplification seems to play an important role for many different drivers.
Zofia Piotrowska, MD, MHS
We will wrap up our discussion of advances in targeted therapy for advanced NSCLC with discussion of what I think is a really exciting study to come out of ASCO 2020: DESTINY-Lung01.
DESTINY-Lung01 is an ongoing, open-label, multicenter, multicohort, phase II study evaluating T-DXd 6.4 mg/kg every 3 weeks in patients with pretreated, metastatic NSCLC that is HER2 expressing or has a HER2-activating mutation (planned N = 170).31 T-DXd is already approved by the FDA for the treatment of HER2-positive metastatic breast cancer,32 and there is precedence that HER2-targeted drugs for breast cancer can also be effective in patients with HER2-mutated NSCLC.
Cohort 1 comprised patients with HER2 overexpression, whereas Cohort 2 comprised those with HER2-mutated NSCLC. Typically, the HER2 mutations were exon 20 insertions, which account for more than 90% of HER2 mutations.33 Although these mutations are found in only approximately 2% of NSCLC cases34, they are emerging as an important genetic driver in the patients who have them.
At ASCO 2020, Smit and colleagues presented data from an interim analysis for the 42 patients with HER2-mutated NSCLC in Cohort 2.35 The primary endpoint was ORR with secondary endpoints including DOR, PFS, and OS. The majority of these patients had received platinum-based therapy (90.5%) and just over one half (54.8%) had received PD-1/PD-L1–targeted treatment.
Zofia Piotrowska, MD, MHS
In patients with HER2-mutated NSCLC, the ORR (by independent central review) to T-DXd was 61.9%, and the median duration of response was not reached.
Zofia Piotrowska, MD, MHS
This impressive waterfall plot shows that all 39 evaluable patients with HER2-mutated NSCLC had a tumor size reduction with T-DXd, suggesting a clinical benefit for this drug in this setting.
Zofia Piotrowska, MD, MHS
This swimmers plot demonstrates that responses occurred early in treatment, often at the first restaging scan and, in some cases, at the second. With the caveat that this is a small group, many of these patients have been on treatment for quite a long time (up to 60 weeks in a few cases), which suggests a potentially durable clinical benefit from T-DXd.
Zofia Piotrowska, MD, MHS
The estimated median PFS with T-DXd was 14 months (95% CI: 6.4-14.0), which I think is clinically meaningful. Median OS was not reached (95% CI: 11.8 to not reached).
Zofia Piotrowska, MD, MHS
T-DXd exhibited toxicities expected to be associated with an antibody–drug conjugate with a chemotherapy “payload,” most commonly nausea and alopecia. However, the majority of T-DXd–associated AEs were grade 1/2. The 2 most common grade ≥ 3 AEs were myelosuppressive, including anemia and decreased neutrophil count.
Zofia Piotrowska, MD, MHS
There were 5 deaths in this study, although none were deemed to be related to treatment by the investigators. A total of 64.3% of patients experienced grade ≥ 3 treatment-emergent AEs, 52.4% of which were deemed drug related. Of note, 23.8% of patients discontinued T-DXd due to a treatment-emergent AE, with an additional 38.1% reducing their dose and 59.5% interrupting their dose. In my opinion, the toxicities from T-DXd must be carefully monitored and weighed against its clinical benefit.
Zofia Piotrowska, MD, MHS
Of importance, 5 patients in DESTINY‑Lung01 developed grade 2 interstitial lung disease (11.9%) with T-DXd, and all of them required steroid treatment. Four discontinued treatment, and of the 5, 1 had not recovered by the data cutoff. Patients receiving T-DXd must be monitored closely for signs and symptoms of interstitial lung disease.
Zofia Piotrowska, MD, MHS
To date, treatment options for patients with HER2‑mutated NSCLC have been limited. There have been preliminary signs of activity with oral TKIs that target both EGFR and HER2 in patients with exon 20 insertions, but outcomes with those drugs have been modest and rates of toxicity significant.36-39 In a phase II study, the HER2-targeted antibody–drug conjugate ado-trastuzumab emtansine (T-DM1) also showed some activity in patients with HER2‑mutated NSCLC, with an ORR (all PRs) of 44%.40
By contrast, I was impressed with the 61.9% response rate and median PFS of 14 months reported from DESTINY-Lung01 for patients with HER2-mutated NSCLC. I think T-DXd is a treatment option with promise for HER2‑mutated NSCLC that warrants further study but one with some notable toxicities.
Stephen V. Liu, MD
I share your enthusiasm. This is a very important dataset for a genomically defined subset of patients for whom there is no approved treatment. Although T-DM1 is not yet approved for HER2-mutated NSCLC, it is listed in clinical guidelines and I’ve used it with some success. However, the PFS with T-DM1 in HER2-mutated NSCLC is only approximately 5 months.40 Here with T-DXd, I was quite impressed with the durability of response and a PFS exceeding 1 year. However, we need to see how these data mature, noting the confidence intervals. These results also highlight that T-DXd is more toxic than other targeted agents, with a fair degree of nausea, alopecia, and myelosuppression, as well as interstitial lung disease, so we’ll want to see how the efficacy and safety profiles play out in a larger set of patients. Overall, I am hopeful we will have T-DXd as an option for patients with HER2-mutated NSCLC in the near future.
Zofia Piotrowska, MD, MHS
Absolutely. Here we have another rare target that we haven’t been broadly looking for, but is something that will be important to identify in our patients as effective therapies become available. So, again, I will highlight the importance of using broad next-generation sequencing, including genes such as HER2, for our patients with newly diagnosed NSCLC, so that we can identify those who may be eligible for these types of therapy.
Back
