CME
Physicians: Maximum of 1.25 AMA PRA Category 1 Credits™
Released: September 25, 2020
Expiration: September 24, 2021
Zofia Piotrowska, MD, MHS
Exciting data were presented at ASCO 2020 regarding adjuvant EGFR tyrosine kinase inhibitor (TKI) therapy for patients with early-stage EGFR-mutated NSCLC. Both the ADAURA and ADJUVANT-CTONG 1104 phase III clinical trials provided promising data regarding improved DFS with adjuvant osimertinib and gefitinib, respectively. We will now discuss both studies, with the caveat that longer-term follow-up will be needed to clarify the clinical role of adjuvant EGFR TKI therapy in early-stage NSCLC.
Zofia Piotrowska, MD, MHS
ADAURA is a large, ongoing, international, randomized, double-blind phase III trial comparing osimertinib—a third-generation EGFR inhibitor—with placebo as adjuvant therapy for adult patients with stage IB-IIIA, EGFR-mutated (either exon 19 deletion or L858R), nonsquamous NSCLC after complete resection.1 Adjuvant chemotherapy was permitted per standard of care at the local institution.
At ASCO 2020, Herbst and colleagues presented exciting results from an unplanned interim analysis of ADAURA conducted due to early indication of efficacy (after which the study was unblinded). Of note, all of the patients had at least 1 year of follow-up at the time of this analysis (cut off: January 2020).
Patients (N = 682) were randomized (at 10 weeks after surgery without adjuvant chemotherapy or 26 weeks with adjuvant chemotherapy) to either the standard dose of osimertinib 80 mg/day (n = 339) or placebo (n = 343) for up to 3 years. Patients were stratified by disease stage, EGFR mutation, and race (Asian vs non-Asian). The primary endpoint was investigator‑assessed DFS in patients with stage II/IIIA disease. Secondary endpoints included DFS in the overall population, DFS at Years 2-5, safety, health-related quality of life, and OS.
Overall, the arms were well balanced, with each arm having approximately one third of patients with stage IB, II, or IIIA disease. The percentage of patients who received adjuvant chemotherapy was also equivalent between the arms (osimertinib 55%; placebo 56%), but overall it was lower than expected for patients with stage I-IIIA disease. Receipt of adjuvant chemotherapy by stage, which could have shed light on this low rate, was not reported.
Zofia Piotrowska, MD, MHS
The primary endpoint—DFS in patients with stage II and IIIA NSCLC—shows good separation of the curves between patients receiving adjuvant osimertinib vs placebo. The median DFS was not reached with osimertinib vs 20.4 months with placebo (HR: 0.17; 95% CI: 0.12-0.23; P < .0001).
Increasing separation of the curves over time can be seen in the landmark analysis. At 12 months, 97% of patients receiving osimertinib were disease free vs 61% on the placebo arm. Likewise, respective DFS rates were 90% vs 44% at 24 months and 80% vs 28% at 36 months, respectively.
Zofia Piotrowska, MD, MHS
With regard to the secondary endpoint of DFS in the overall study population (stage IB-IIIA), the median was not reached with osimertinib vs 28.1 months with placebo (HR: 0.21; 95% CI: 0.16-0.28; P < .0001). At 12 months, 97% of patients in the osimertinib arm were disease free vs 69% on placebo. At 24 months, these rates were 89% vs 53%, and at 36 months, they were 79% vs 41%, respectively.
Of note, these DFS data are only approximately 30% mature; however, even in this early analysis, a consistent separation in DFS that increases over time is seen.
Zofia Piotrowska, MD, MHS
As shown here, the magnitude of DFS benefit with adjuvant osimertinib clearly increased with worsening disease stage. The benefit was relatively modest in patients with stage IB disease, with 2-year DFS rates of 87% vs 73% with placebo (HR: 0.50; 95% CI: 0.25-0.96). In striking contrast, patients with higher-risk stage IIIA disease had much greater benefit, with 2-year DFS rates of 88% vs 32%, respectively (HR: 0.12; 95% CI: 0.07-0.20).
Although these DFS rates are impressive, is demonstration of a DFS benefit with adjuvant therapy enough to change clinical practice? Rather, what we would really like to see for an adjuvant study to be practice changing is an OS benefit.
Zofia Piotrowska, MD, MHS
In this interim analysis, the median OS was not reached in either arm and the OS curves very closely aligned, showing high rates of survival in both arms (HR: 0.40; 95% CI: 0.18-0.90). However, these OS data are very immature (5% maturity) and not yet statistically significant, so there is great interest in the longer follow-up data for this study.
Zofia Piotrowska, MD, MHS
In ADAURA, the safety profile of osimertinib, which is generally well tolerated, was consistent with previous reports. The median duration of osimertinib exposure was 22.3 months (range: 0-43), which suggests that patients were able to stay on this drug for up to 3 years, as originally planned. Of the patients in the osimertinib arm, 11% discontinued treatment and 7% reduced the dose due to an AE. This is notable compared with older studies of adjuvant EGFR TKIs, such as erlotinib, which demonstrated shorter median durations of exposure because those drugs had higher rates of toxicity and, therefore, more patients discontinued early.2
As expected, osimertinib-associated toxicities were seen, including low-grade diarrhea (46 patients vs 19 in the placebo arm), paronychia (25 vs 1), and dry skin (23 vs 6). Generally, however, rates of high‑grade toxicities were quite low, which is consistent with osimertinib being a well-tolerated drug.
Of note, there was a 3% rate of grade 1/2 interstitial lung disease for patients receiving osimertinib, which is consistent with rates in the metastatic setting.3 In addition, receipt of osimertinib was potentially related to corrected QT prolongation (another known side effect of osimertinib) in 7% of patients.
Zofia Piotrowska, MD, MHS
Dr. Liu, what are your thoughts on the ADAURA data? Do you think adjuvant osimertinib is ready for use in the clinic for our patients with early-stage, EGFR-mutated NSCLC based on the DFS benefit revealed so far?
Stephen V. Liu, MD
Ultimately, I do think ADAURA is a practice‑changing study, despite not being a perfect dataset. As discussed, flaws include that 45% of patients did not receive adjuvant chemotherapy, a percentage that seems high based on current practice, although it was balanced between the arms. Approximately 30% of patients had stage IB disease, so hopefully, the patients with stage II/III disease comprise the majority of those who received adjuvant chemotherapy. Other flaws in this dataset include the lack of postoperative radiation therapy, which is currently my standard treatment for patients with resected N2 NSCLC. Also, preoperative PET scans were not mandated for staging. Although central nervous system (CNS) imaging was used, investigators did not mandate the modality; magnetic resonance imaging is optimal in that setting. That said, flaws are the realities of any global trial, and the sheer magnitude of DFS benefit for osimertinib shown in ADAURA cannot be ignored.
Zofia Piotrowska, MD, MHS
You bring up some important key limitations of these data, but as mentioned, this was a very early look, and we certainly will need a lot more granularity in the long term to be able to take this data to the clinic.
It is noteworthy that these results build on previous, smaller studies of adjuvant EGFR TKI therapy in EGFR-mutated NSCLC that consistently have shown a DFS benefit, but have not yet shown an OS benefit.2,4 Delaying time to disease recurrence is certainly of clinical importance, but our ultimate goal is cure in the early-stage setting and the OS results of ADAURA will be essential to see if adjuvant osimertinib is truly curative. However, mature OS data won’t be available for years, so in the meantime, once osimertinib receives approval by the FDA in this setting, clinicians are going to have to rely on these impressive DFS HRs to make decisions about whether to prescribe adjuvant osimertinib. In my opinion, these data are impressive and certainly do merit consideration of adjuvant osimertinib in careful discussion with our patients.
Stephen V. Liu, MD
Yes, this is a perfect setting for shared decision making, where we present the ADAURA data to our patients and have informed conversations with them about the pros and cons of adjuvant osimertinib therapy.
I also think it is very likely that the DFS benefit will translate into an OS benefit but that it might not necessarily mean we are curing more patients. The patients in ADAURA were on treatment for 3 years, and I expect the DFS curves to stay apart for some time after stopping therapy; but what if the curves come back together at Year 4 or 5? If we do see those curves come together at some point after stopping treatment, it might mean that some patients were experiencing delayed relapse instead of a cure.
Furthermore, there are downsides of giving osimertinib in the adjuvant setting. One of these is the potential for overtreatment, in particular for patients with stage IB disease that already may be cured. We don’t want to give patients treatment they don’t need. Moreover, although osimertinib is a fairly tolerable drug, more than 10% of patients discontinued treatment in ADAURA due to an AE,1 so toxicities and quality-of-life issues must be considered. Realistically, however, the major drawback is cost. Some patients have high insurance copays that, over 3 years (or longer), could represent a substantial financial burden for the patient and their family, not to mention the societal cost of using a fairly expensive medication for 3 years when they may not have needed it.
Zofia Piotrowska, MD, MHS
That is a good point. If it turns out that we are just delaying time to recurrence, maybe not all patients should receive osimertinib as adjuvant therapy. Perhaps they would receive a similar benefit if osimertinib was given only at the time of recurrence? We don’t yet have an answer to this question.
We also don’t know whether patients who progress following adjuvant osimertinib can benefit from osimertinib in the metastatic setting. Hints from earlier studies suggest that the answer may be yes. For example, in the single‑arm phase II SELECT study of adjuvant erlotinib for 2 years after standard adjuvant chemotherapy (with or without radiation) in patients with EGFR-mutated NSCLC, 65% of the 40 patients who suffered disease recurrence experienced durable benefit when rechallenged with erlotinib.4 Again, it is hard to extrapolate efficacy with osimertinib from that finding—these are important questions that will require longer follow-up from ADAURA to answer.
Stephen V. Liu, MD
Another unanswered question is which patients would most benefit from adjuvant osimertinib. My hope is that, in the years to come, the real legacy from ADAURA will be in the correlative studies; that there may be a biomarker that identifies those patients with early-stage disease destined for relapse who would strongly benefit from adjuvant osimertinib.
Zofia Piotrowska, MD, MHS
ADJUVANT-CTONG 1104 is a randomized, open‑label, phase III study of adjuvant gefitinib vs chemotherapy in patients with resected EGFR-mutated NSCLC.5-7 The study was conducted at multiple centers in China with patients enrolled between 2011 and 2014.
Patients (N = 222) with resected stage II‑IIIA (N1-N2) NSCLC who had an EGFR exon 19 deletion or a L858R mutation were randomized 1:1 to either the first-generation EGFR inhibitor gefitinib 250 mg/day for 24 months or to chemotherapy with cisplatin (75 mg/m2 on Day 1 every 2 weeks for 4 cycles) and vinorelbine (25 mg/m2 on Days 1, 8, and up). The primary endpoint was DFS. Secondary endpoints included OS (including the 5-year rate), the 3-year and 5-year DFS rate, health-related quality of life, an exploratory biomarker analysis, and safety.
Results from CTONG 1104 showing superior DFS with gefitinib were presented at the 2017 ASCO Annual Meeting.6 The analysis that was presented at ASCO 2020 included final OS results and longer-term (3-year and 5-year) follow-up results for DFS in both the ITT population and the per-protocol population.7
Zofia Piotrowska, MD, MHS
Updated results showed a DFS benefit in both the ITT and per-protocol populations; for both groups the median DFS was 30.8 months with gefitinib vs 19.8 months for chemotherapy, with an HR of 0.56 (95% CI: 0.40-0.79; P = .001) and 0.51 (95% CI: 0.36-0.72; P < .001) for the ITT and per protocol groups, respectively.
In both groups, the 2 curves separated early and continued getting farther apart until approximately the end of gefitinib therapy at 24 months and then rejoined around Month 48. These results suggest, as we discussed above, that the therapeutic benefit is magnified during treatment, and that treatment may only be delaying time to recurrence rather than truly curing these patients.
Zofia Piotrowska, MD, MHS
The OS data, an eagerly awaited outcome of this study, were disappointing. The median OS was prolonged, but not significantly improved with gefitinib. As shown here, the OS benefit with gefitinib of just 12.3 months was the same for both the ITT group and the per protocol group, with an HR of 0.92 for both. In addition, subgroup analyses (age, sex, EGFR mutation, nodal status) revealed no group with significant benefit from adjuvant gefitinib.
Zofia Piotrowska, MD, MHS
I think the results reported for ADJUVANT-CTONG 1104 are noteworthy in the context of the ADAURA trial. In particular, it is important to note that CTONG 1104 randomized patients to either an adjuvant EGFR inhibitor or to adjuvant chemotherapy. Because we know there is an OS benefit with adjuvant chemotherapy for patients with stage II/IIIA disease,8 it is difficult to extrapolate the CTONG 1104 results to clinical practice because adjuvant chemotherapy was not administrated uniformly to all enrolled patients. In fact, it’s hard to even interpret the OS data being negative in this study without the uniform use of adjuvant chemotherapy in both arms. In contrast, the design of ADAURA allowed patients to get adjuvant chemotherapy before randomization to adjuvant EGFR TKI therapy vs placebo, which makes the ADAURA results more readily applicable to current practice.
Stephen V. Liu, MD
I think the results of CTONG 1104 also might provide an important lesson: A substantial improvement in DFS does not guarantee an improvement in OS. Just as you alluded to, Dr. Piotrowska, separation of the curves may only occur during treatment. There are different ways to interpret that: Does it mean there is no meaningful benefit from adjuvant therapy? Or does it imply that treatment needs to be extended?
Either way, it’s hard not to worry a little bit that the CTONG 1104 results might be giving us a clue as to what we may see with long-term follow-up for ADAURA. It will be disappointing if in 5 to 10 years from now the ADAURA OS curves start to match those in CTONG 1104.
Zofia Piotrowska, MD, MHS
Good point. However, I do have to say that with adjuvant chemotherapy having been given only in the control arm in CTONG 1104, I am hopeful that we won’t see the same trend with longer-term follow-up of ADAURA, where adjuvant chemotherapy was allowed in both arms. Although the rates of adjuvant chemotherapy administration were lower than anticipated in ADAURA, as we discussed, they were balanced across the arms. So I do wonder, with longer follow-up, whether that uniform use of chemotherapy in both arms of ADAURA will impact OS outcomes or whether it really will be that it’s more of a matter of patients having an improvement in DFS while they’re on osimertinib that lasts only while they’re on treatment and then washes away with longer follow-up.
Stephen V. Liu, MD
You are right in that there are some important differences between these 2 studies, and that the chemotherapy imbalance in CTONG 1104 is glaring. There is also concern that some of the patients in CTONG 1104 were understaged. If some patients did indeed have occult metastatic disease, we could expect an EGFR TKI to offer DFS benefit but not OS benefit. Osimertinib is more effective, is better tolerated, and has better CNS penetration than gefitinib, which might be of particular importance in this context.
Zofia Piotrowska, MD, MHS
Another interesting question that I’ve been asked a few times since ADAURA was presented at ASCO 2020 is whether the data suggest that we can skip adjuvant chemotherapy for patients who have a known EGFR mutation in favor of using adjuvant osimertinib. Because of the demonstrated lack of OS benefit with adjuvant EGFR TKI therapy compared with adjuvant chemotherapy in the CTONG 1104 study, I would caution clinicians against skipping adjuvant chemotherapy, unless a patient was ineligible. We can certainly consider whether to use adjuvant osimertinib following chemotherapy, but not instead of an approach with a proven survival benefit.
Back
