ASCEND-7: Efficacy and Safety of Ceritinib Treatment in Patients with ALK-Positive Non–Small Cell Lung Cancer Metastatic to the Brain and/or Leptomeninges

This article is featured in Highlights of This Issue, p. 2475

Central nervous system (CNS) metastases frequently occur in patients with anaplastic lymphoma kinase-positive (ALK⁺) non–small cell lung cancer (NSCLC) and have a negative impact on quality of life and survival (1). About 24% to 30% of patients with ALK⁺ NSCLC present with brain metastases at the time of initial diagnosis (2, 3) and about half develop brain metastases at some point during their illness (3, 4). Leptomeningeal metastases are found in about 5% to 6% of patients with ALK⁺ NSCLC and usually present as a late complication (5).

Despite considerable advances in the treatment of NSCLC in recent years, the prognosis remains poor for patients with brain and leptomeningeal metastases (6–8). Treatments for brain metastases—usually palliative, aiming to achieve local control of the CNS disease—include surgical resection, stereotactic radiosurgery, or whole-brain radiation therapy (WBRT). The median survival for patients with NSCLC with brain metastases is approximately 1 to 3 months in the absence of treatment and approximately 5 to 7 months with WBRT (9–12). For patients with leptomeningeal metastases, the median survival without treatment is approximately 4 to 6 weeks and approximately 4 months with available treatments (WBRT and intrathecal chemotherapy; refs. 13, 14).

Crizotinib has shown modest intracranial activity and poor blood–brain barrier (BBB) penetration (15). Acquired resistance to crizotinib develops usually within 1 to 2 years in responders, with CNS metastases being a frequent site of progression, where up to 60% of patients develop metastasis (16–20). ALK inhibitors (ALKi), including alectinib, brigatinib, lorlatinib, crizotinib, and ceritinib, have demonstrated clinical activity in patients with brain metastases (21–25). However, eligibility criteria and CNS assessments were inconsistent across studies, which limit a clear understanding of the CNS activity of next-generation ALKis. In most trials, the assessment of intracranial disease was a secondary endpoint. As patients recently treated with brain radiation were also included in these trials, the intracranial activity of the ALKi could have been confounded by brain radiation. Patients with active metastases (untreated) are often ineligible for, or underrepresented in, clinical trials of systemic therapies (26, 27). No previous trials have been conducted specifically to study the CNS effects of targeted ALKis.

Ceritinib, an orally available second-generation ALKi, is approved for the treatment of patients with ALK⁺ advanced/metastatic NSCLC (28, 29). In enzymatic assays, ceritinib is approximately 20-fold more potent than crizotinib (29). It inhibits phosphorylation of ALK and downstream PI3K/AKT, MEK/ERK, and mTOR signaling pathways (29). In rat xenograft models, ceritinib inhibited EML4-ALK, resulting in inhibition of tumor growth and tumor regression (29). It has also shown potent antitumor activity against crizotinib-resistant H2228 NSCLC cell lines, including resistant variants carrying I1171T or C1156Y mutations in the ALK kinase domain (29).

It has shown promising intracranial activity in patients who were treatment-naïve or pretreated with crizotinib (30, 31). Although other ASCEND studies did include patients with brain metastasis and have established the safety and efficacy of ceritinib, ASCEND-7 is the first study specifically designed to study the effect of ceritinib in patients with ALK⁺ NSCLC metastatic to brain and/or leptomeninges. The ASCEND-7 study (NCT02336451) was designed in 2015 at a time when ceritinib was among the first ALKis to demonstrate efficacy in patients who had progression of disease on crizotinib treatment. This study only included patients with ALK⁺ NSCLC (as determined using the FDA approved Vysis ALK Break Apart FISH Probe Kit [Abbott Molecular Inc.] test and scoring algorithm, including positivity criteria, if locally available or at a Novartis designated central laboratory) with active brain metastases and/or leptomeningeal carcinomatosis (LC), symptomatic or not, who were stratified by prior treatment to assess the impact of prior brain radiation or prior exposure to ALKis on the intracranial efficacy of ceritinib.

ASCEND-7 was a phase II, multicenter, global (conducted in 17 countries), open-label, 5-arm study assessed the efficacy and safety of oral ceritinib. Eligible patients (aged ≥ 18 years) had ALK⁺ NSCLC with active metastases into the brain and/or to leptomeninges confirmed by gadolinium-enhanced magnetic resonance imaging (MRI) at the time of study entry; ≥ 1 measurable extracranial lesion as defined by the RECIST v1.1, with or without neurological symptoms; and World Health Organization performance status of 0 to 2. Prior treatment with chemotherapy was permitted. Patients were allocated to arm 1 to 4 depending on their history of prior therapy (prior radiotherapy to the brain and prior treatment with an ALKi) or to the exploratory arm 5 if the patients had evidence of LC (Fig. 1).

Following completion of screening procedures and verifying patient eligibility, patients were allocated to one of the arms 1 to 4 if they had brain metastases or to arm 5 if they had LC, according to their prior treatment based on investigator assessment at baseline.

• Arm 1: Patients previously treated with radiation to the brain and with prior exposure to an ALKi.

• Arm 2: Patients previously untreated with radiation to the brain but with prior exposure to an ALKi.

• Arm 3: Patients previously treated with radiation to the brain but with no prior exposure to an ALKi.

• Arm 4: Patients previously untreated with radiation to the brain and with no prior exposure to an ALKi.

• Arm 5: Patients with LC

An active brain lesion was defined as a lesion free of any local treatment (e.g., stereotactic radiosurgery or WBRT; see Supplementary Data for details).

The study was approved by the ethics committee and/or institutional review board for each center and was conducted in accordance with the ethical principles laid down in the Declaration of Helsinki and the guidelines for Good Clinical Practice. All patients provided written informed consent before screening.

Eligible patients were treated with ceritinib 750 mg/day orally, in a fasted state (fast from food and drink, except water, at least 1 hour before or 2 hours after a light meal). Treatment with ceritinib was continued until disease progression, withdrawal of consent, or discontinuation at the discretion of the investigator.

The primary endpoint was investigator-assessed overall response rate (ORR), defined as the proportion of patients with a best overall confirmed response of complete response (CR) or partial response (PR) in the whole body (combination of intracranial and extracranial responses). The key secondary endpoint was whole-body disease control rate (DCR) per investigator assessment. Other secondary endpoints included ORR, DCR, time to response, and duration of response (DOR), respectively, in the brain (intracranial), outside of the brain (extracranial), and in the whole body (for lesions in and outside the brain), along with whole-body progression-free survival (PFS), and overall survival (OS). Further details on assessments are included in the Supplementary Data.

The whole-body ORR was estimated, and the exact binomial 95% confidence intervals (CI) were provided by study arm. Whole-body DCR and intracranial/extracranial ORR and DCR were estimated, and the associated exact binomial 95% CI was reported. Duration of response, PFS, and OS were analyzed using the Kaplan–Meier method to estimate the median value. All statistical analyses were performed using SAS v9.4.

Novartis is committed to sharing with qualified external researchers, access to patient-level data and supporting clinical documents from eligible studies. These requests are reviewed and approved by an independent review panel on the basis of scientific merit. All data provided are anonymized to respect the privacy of patients who have participated in the trial in line with applicable laws and regulations. This trial data availability is according to the criteria and process described on www.clinicalstudydatarequest.com.

From April 2015 to February 2019, 246 patients were screened and 156 patients met eligibility to receive treatment after being assigned to one of the study arms (42 in arm 1; 40 in arm 2; 12 in arm 3; 44 in arm 4; and 18 in arm 5). Enrollment of patients with brain metastasis who had received prior radiotherapy to the brain but no prior ALKi was difficult, explaining the lower number of patients included in arm 3 compared with other arms. The median duration of study follow-up (from start date of study drug to data cutoff date) was 33.84 months (interquartile range, 29.55–37.52; n = 156). At the time of data cutoff (February 06, 2019), patients still benefiting from ceritinib treatment were transitioned to rollover study (n = 31) or continued treatment with commercially available drug (n = 2) or managed access program (n = 1). The most frequent reason for treatment discontinuation was progressive disease, reported in 69 patients (44.2%; Supplementary Table S1).

The median age ranged from 46.0 to 53.5 years across the arms, and most patients (50%-74%) were never smokers. All patients in arm 1 (prior brain RT/prior ALKi), most patients in arm 2 (no prior brain RT/prior ALKi, 97.5%) and arm 5 (patients with LM, 88.9%) had received prior crizotinib. Most patients enrolled in the ASCEND-7 study had received prior chemotherapy (Table 1). The median intracranial disease burden (sum of diameters for intracranial target lesions) at baseline based on Investigator assessment was 25 mm (10.0 –159.0). Most frequent location of these intracranial target lesion were frontal in 23.7% of the patients, cerebellar in 19.9% of the patients, parietal in 16.7% of patients, occipital in 15.4% of patients, and temporal in 14.7% of patients. Similar distribution pattern is seen across non-target lesions as assessed by the Investigator.

The investigator-assessed whole-body ORR was higher in arm 4 (no prior brain RT or ALKi, 59.1%; 95% CI, 43.2–73.7) compared with the other arms: 35.7% (95% CI, 21.6–52.0) in arm 1 (prior brain RT/prior ALKi), 30.0% (95% CI, 16.6–46.5) in arm 2 (no prior brain RT/prior ALKi), and 50.0% (95% CI, 21.1–78.9) in arm 3 (prior brain RT/no prior ALKi; Table 2; Fig. 2). The whole-body ORR was higher in patients with no prior exposure to an ALKi (combined arms 3 + 4, n = 56) than in patients with prior exposure to an ALKi (combined arms 1 + 2, n = 82; 57.1%; 95% CI, 43.2–70.3 vs. 32.9%; 95% CI, 22.9–44.2). The whole-body ORR was similar in the group of patients with no prior brain radiotherapy (combined arms 2 + 4, n = 84) compared with the group of patients with a brain radiotherapy (combined arms 1 + 3, n = 54) with overlapping CIs; 45.2% (95% CI, 34.3–56.5) and 38.9% (95% CI, 25.9–53.1), respectively (Supplementary Table S2). The BIRC results were generally consistent with the investigator assessment (Supplementary Table S3).

Investigator-assessed whole-body DCR was high in all arms but highest in arm 2 (no prior brain RT/prior ALKi; Table 2). The whole-body DCR was similar in patients regardless of previous ALKi therapy and radiotherapy (Supplementary Table S2,). Whole-body median DOR was similar in patients with or without prior exposure to ALKi and in patients who did or did not receive prior brain radiation. However, considering the low number of responders, results must be interpreted cautiously. Whole-body median DOR values for each arm are listed in Table 2. The median PFS based on investigator assessment was 7.2 months (95% CI, 3.3–10.9), 5.6 months (95% CI, 3.6–9.2), 7.9 months (95% CI, 5.5–9.4) in arms 1, 2, and 4. Median PFS was not reached in arm 3 patients. The 12-month PFS rate was 66.7% (95% CI, 33.7%–86.0%) in arm 3. Median PFS was 6.9 months (95% CI, 3.7–7.4) in patients with prior ALKi use (arms 1 + 2) and 9.1 months (95% CI, 5.6–12.7) in patients without prior ALKi use (arms 3 + 4). The OS data was not mature and must be interpreted with caution. The median OS was 24.0 (95% CI, 12.6–NE) months, and 28.5 (95% CI, 21.0–NE) months in arms 1 and in all patients. OS was not estimable in arms 2 to 4, respectively. Kaplan–Meier estimates of median PFS and OS for each arm are presented in Supplementary Figs. S1 and S2, and PFS values as per previous therapy are provided in Supplementary Materials (see Supplementary Data for details).

Intracranial ORR based on investigator assessment as per modified RECIST 1.1 was higher in arm 4 (no prior brain RT or ALKi) compared with arms 1 to 3 (Table 3; Fig. 2). Intracranial DCR at weeks 8 and 16 was high in all arms. Intracranial ORR was higher in patients who were ALKi-naïve (47.5%; 95% CI, 31.5–63.9 in combined arms 3 + 4) compared with patients who received prior ALKi (33.3%; 95% CI, 21.4–47.1 in combined arms 1 + 2). Intracranial response was similar in patients who received radiation (37.1%; 95% CI, 21.5–55.1 in combined arms 1 + 3) and those who did not (40.3%; 95% CI, 28.1–53.6 in combined arms 2 + 4). Regression of tumors based on prior ALKi therapy was also evident from the corresponding waterfall plot (Fig. 3).

The swimmer's plot in Fig. 4 shows the treatment duration of all five arms. The median time to intracranial response per investigator assessment was 1.87 months (range, 1.7 7.5) in arm 1, 1.84 months (range, 1.6 9.1) in arm 2, 3.56 months (range, 1.8 5.3) in arm 3, and 1.77 months (range, 1.3 7.4) in arm 4. Although the median value was higher in arm 3 compared with other arms, it must be noted that only two patients were included in the assessment. Extracranial ORR was higher in arm 4 compared with arms 1 to 3 (Supplementary Table S4).

Disease progression occurred most frequently in the brain. Patients who had not received brain radiotherapy had more frequent progression. Five patients in arm 1; 4 in arm 2; none in arm 3; 4 in arm 4; and 1 in arm 5 had whole-body disease progression that included both intracranial and extracranial compartments.

In arm 5 (leptomeningeal disease, n = 18), whole-body ORR was 16.7% (95% CI, 3.6–41.4), whereas whole-body DCR was 66.7% (95% CI, 41.0–86.7). Intracranial ORR in patients with LC presented with concomitant measurable brain metastases at baseline (n = 8) was 12.5% (95% CI, 0.3–52.7); extracranial ORR was 22.2% (95% CI, 6.4–47.6). One patient with LC had an unconfirmed PR. The median PFS in patients with LC was 5.2 months (95% CI, 1.6–7.2), and the median OS was 7.2 months (95% CI, 1.6–16.9).

Steady-state drug concentration was reached after 4 weeks of daily dosing and remained stable afterward, and was consistent with previously reported studies in at the 750-mg once-daily dose (Supplementary Fig. S3; ref. 32). Paired cerebrospinal fluid (CSF) and plasma sampling from three patients revealed that ceritinib penetrated the human BBB with a CSF-to-plasma concentration ratio ranging from 13% to 35% (Supplementary Table S5).

An overview of adverse events (AE) is provided in Supplementary Table S6. The most frequent AEs (regardless of study drug relationship, ≥ 20% of all patients, Supplementary Table S7) were diarrhea (68.6%), nausea (55.8%), increased alanine aminotransferase (ALT; 47.4%), vomiting (46.8%), increased aspartate aminotransferase (35.9%), and decreased appetite (32.1%). Grade 3 or 4 AEs (regardless of study drug relationship) were reported in 123 (78.8%) of the 156 patients; the most frequent (>10% of all patients) were increased ALT (28.2%) and increased gamma-glutamyltransferase (16.7%).

Overall, 31 on-treatment deaths (19.1%; while receiving ceritinib or within 30 days of last dose) were reported. Details on deaths by treatment arm are mentioned in Supplementary Materials.

Ceritinib has demonstrated clinically meaningful intracranial/extracranial tumor responses in patients with ALK⁺ NSCLC metastatic to the brain and/or leptomeninges. Intracranial/extracranial responses were higher in patients with no prior exposure to an ALKi, as expected but prior brain radiation did not seem to influence the tumor response to treatment. These results confirm previously reported activity of ceritinib in patients with ALK⁺ NSCLC with CNS metastases in patients naïve to an ALKi or pretreated with crizotinib (30, 31). In asymptomatic or minimally symptomatic ALK⁺ NSCLC patients with CNS involvement, second- and third-generation ALK TKIs, including ceritinib, with high intracranial activity and robust anti-tumor efficacy may be recommended as front-line therapy with deferral of local treatment such as radiation (21, 33, 34). In highly symptomatic patients or those with mass effect or herniation, radiation or surgery should be considered upfront.

Although WBRT may be used for control of CNS disease, it is commonly deferred where possible due to its toxicity and cognitive neurologic impacts. At the time of study conception, the ability of targeted agents to affect CNS disease was not well elucidated. Although crizotinib is effective in ALK⁺ NSCLC, its benefit is limited due to poor BBB penetration and early disease development and progression intracranially (16–20, 35).

At the time of study design, several second-generation and third-generation ALKis demonstrated reports of benefit in patients with ALK⁺ NSCLC with brain metastases but the inclusion criteria across studies varied considerably in terms of symptomatic/asymptomatic brain metastases, limiting inclusion to only controlled brain metastases, or permitted prior treatments (radiotherapy and chemotherapy, as well as ALKi therapy). ASCEND-7 is the first study designed to specifically assess and define the efficacy of ceritinib on intracranial and extracranial disease with homogenous well-defined eligibility criteria in patients with active brain metastases. Most patients enrolled in the ASCEND-7 study had received prior chemotherapy.

In ASCEND-7, the higher whole-body ORR in arm 4 (no prior brain RT or ALKi) and in combined arms 3 + 4 (ALKi-naïve patients) suggests that patients with active brain metastases who were ALKi-naïve achieved higher response than patients with a prior exposure to an ALKi. Extracranial and intracranial ORRs were also higher in arm 4. DCR was high across all patients with brain metastases treated in this study irrespective of prior treatment with ALKi or prior brain radiation.

Ceritinib has demonstrated intracranial activity in other trials. An intracranial ORR of 72.7% was recorded in treatment-naïve patients with measurable brain metastases at baseline treated with ceritinib in ASCEND-4 (30). In patients without prior ALKi use (arms 3 + 4) in ASCEND-7, intracranial ORR was 47.5% months. In the ASCEND-5 study, the intracranial ORR was 35% in crizotinib-pretreated patients with measurable baseline brain metastasis, whereas the ASCEND-7 study reports an intracranial ORR of 33.3% in patients with prior ALKi use (arms 1 + 2; ref. 31). Intracranial responses in the ASCEND-4 and ASCEND-5 studies were assessed according to modified RECIST 1.1 as assessed by the BIRC neuroradiologist. In a meta-analysis including subgroups of 7 ceritinib studies, the intracranial ORR with ceritinib was 56.6% in ALKi-naïve patients and 41.5% in ALKi-pretreated patients (36).

Recently, other ALKis have demonstrated intracranial response, such as the intracranial ORR of 81% with alectinib in ALKi-naïve patients with measurable brain metastases at baseline (n = 21) in the ALEX study; however, previous radiotherapy was not reported (23). In patients with crizotinib-refractory disease with measurable brain metastasis at baseline, the intracranial ORR with alectinib was 64%. Most of the patients evaluated (34 of 50) had received prior radiotherapy (21).

Newer agents in subsequent studies have also demonstrated intracranial activity. In the phase II ALTA trial in patients with crizotinib-refractory ALK⁺ NSCLC, intracranial ORR and intracranial median PFS with brigatinib were 50% and 12.8 months in arm A (90-mg dose) and 67% and 18.4 months in arm B (180-mg dose), respectively (22). The intracranial ORR with lorlatinib was 87% in patients with brain metastases who had received prior crizotinib, 56% in patients who had received non-crizotinib ALKis, 63% in patients who had received ≥ 1 previous ALKi, and 66.7% in treatment-naïve patients (24). Intracranial ORR was 70% with ensartinib for 40 patients with crizotinib-resistant, ALK⁺ NSCLC with measurable brain metastasis at baseline (25). Most of these recent trials excluded symptomatic patients (21–25). Although newer-generation ALKis have demonstrated intracranial activity, the overall lack of homogeneity in selecting patients with active brain metastases across these trials still makes it difficult to draw any definitive conclusions regarding the efficacy of ALKis independent from prior radiation.

Extracranial ORR was higher in patients who were ALKi-naïve and in patients who did not receive brain radiotherapy compared with patients who had received prior ALKi or prior brain radiotherapy. The extracranial ORR observed in ALKi-pretreated patients in ASCEND-7 (36.6%) was comparable with that reported in patients treated with lorlatinib (37%). However, it is challenging to put the extracranial response data in perspective, because most other trials of ALK inhibitors do not report extracranial response as a separate endpoint but instead rely on whole-body assessments (37).

Although limited CSF pharmacokinetics (PK) data were available and a large degree of interpatient variability in concentrations was observed, paired CSF and plasma sampling revealed that ceritinib crosses the human BBB. This finding is supported by earlier tissue distribution studies of [¹⁴C] ceritinib in rats, which showed that ceritinib is brain penetrant, with a tissue-to-blood exposure (AUC_0-inf) ratio of approximately 15%. Furthermore, correlation between CSF drug exposure and the reported in vitro IC₅₀ of ceritinib, determined in cell-based assay (38), suggests that pharmacologically active drug concentrations are achieved in the brain. The penetration of the BBB is supported by the response observed in patients with leptomeningeal disease. In the 18 patients with ALK⁺ NSCLC metastatic to leptomeninges, most patients (16 of 18) had progression of disease on crizotinib treatment, with crizotinib being the last treatment in 14 patients. Although the whole-body ORR was low (16.7%), the whole-body DCR was relatively high (66.7%). The median OS (7.2 months) in ASCEND-7 was longer than previously reported values in this patient population (13, 14). The median OS after diagnosis of leptomeningeal metastases in unselected patients with NSCLC has been reported to be 4.3 months (39). There are currently no widely accepted response criteria, and the use of RECIST probably underestimates the response rate. Tumor assessments every 4 to 6 weeks may have allowed better assessment of the efficacy of ceritinib in this fragile population of patients. Intracranial progression, alone or with extracranial progression, was the most frequent pattern of progression across arm 1 to 4 but more frequent in patients with no history of prior brain radiation. Although brain radiation plays a role in intracranial disease control, the role of prior brain radiation should be interpreted with caution because patients included across the 4 brain metastasis arms of this study differed in their disease treatment history; some having had more prior treatments.

Each of the currently available ALKis differs substantially in their toxicity profiles. Studies have reported that patients on ceritinib had more gastrointestinal and hepatic toxicities, whereas crizotinib showed more visual disorders and dysgeusia, alectinib showed more dysgeusia, and brigatinib was associated with more respiratory complications (40). The observed safety profile of ceritinib 750 mg administered in fasted conditions in this study is consistent with the known safety profile of ceritinib. No new safety concerns were observed in this study, although higher incidences of nausea and vomiting could have been expected in patients with brain metastases. Dose reductions and dose interruptions due to AEs (seen in 79.5% patients) were common at the 750-mg fasting dose in this trial. The safety, PK, and efficacy data from a dose-optimization study in patients with ALK⁺ NSCLC (ASCEND-8) have demonstrated that ceritinib 450 mg with food has comparable PK and efficacy relative to ceritinib 750-mg fasted, in addition to having a more tolerable toxicity profile; this established the new recommended dose for ceritinib as 450 mg/day taken with food (41, 42).

In conclusion, ceritinib treatment achieved clinically meaningfully high and durable response within and outside the brain in patients with ALK⁺ NSCLC metastatic to the brain and/or leptomeninges, with higher responses in patients who were ALKi-naïve. The intracranial DCR was high irrespective of prior ALKi exposure. As this study confirmed ceritinib has antitumor activity against treated and un-treated brain metastases in patients with ALK⁺ NSCLC with active brain metastases and/or leptomeningeal disease, and could be considered in the management of intracranial disease.

L.Q.M. Chow reports grants, personal fees, and nonfinancial support from Novartis during the conduct of the study as well as grants and personal fees from Alkermes, Merck, AstraZeneca, Pfizer, and Dynavax; grants from Oncorus, Lily/Imclone, Bristol Myers Squibb, Seattle Genetics, and Genentech; and personal fees from Sanofi-Genzyme, EMD Serono, Daiichi Sankyo, Ipsen, Nanobiotix/Life Sciences Dynamic, Regeneron, Blueprint Therapeutics, Beigene, Elicio Therapeutics, Gilead, and Cullinan-Apollo Corp outside the submitted work. F. Barlesi reports personal fees from AstraZeneca, Bayer, Bristol-Myers Squibb, Boehringer-Ingelheim, Eli Lilly Oncology, F. Hoffmann-La Roche Ltd, Novartis, Merck, Mirati, MSD, Pierre Fabre, Pfizer, Seattle Genetics, and Takeda outside the submitted work. E.M. Bertino reports personal fees from Pfizer and Dava Oncology during the conduct of the study as well as personal fees from Pfizer, Intellosphere, and Bristol Myers Squibb, and nonfinancial support from Merck and Lilly outside the submitted work. H.A. Wakelee reports grants from Novartis during the conduct of the study as well as grants from Arrys Therapeutics, Celgene, Clovis, Genentech/Roche, Merck, Pfizer, Exelixis, Gilead, and Pharmacyclics, and grants and personal fees from Xcovery and Mirati outside the submitted work. P.Y. Wen reports other support from Agios, AstraZeneca/Medimmune, Beigene, Celgene, Eli Lilly, Genentech/Roche, Kazia, MediciNova, Merck, Novartis, Nuvation Bio, Oncoceutics, Vascular Biogenics, VBI Vaccines, AstraZeneca, Bayer, Boston Pharmaceuticals, CNS Pharmaceuticals, Elevate Bio Immunomic Therapeutics, Imvax, Karyopharm, Voyager, QED, Celularity, and Sapience outside the submitted work. C.-H. Chiu reports personal fees from Novartis during the conduct of the study as well as personal fees from AstraZeneca, Boehringer-Ingelheim, Bristol-Myers Squibb, Chugai Pharmaceutical, Merck Sharp & Dohme, Eli Lilly, Ono Pharmaceutical, Pfizer, Roche, and Takeda outside the submitted work. M. Majem reports grants and personal fees from Roche, AstraZeneca, and BMS, and personal fees from Sanofi, Janssen, Boehringer-Ingelheim, and Novartis outside the submitted work. M. McKeage reports grants from University of Auckland during the conduct of the study. P. Garrido reports personal fees from AbbVie, Amgen, AstraZeneca, Bayer, Boehringer-Ingelheim, BMS, GSK, Janssen, MSD, Medscape, Novartis, Pfizer, Roche, Takeda, and TouchTime outside the submitted work. F.K. Hurtado reports being an employee of Novartis Pharmaceuticals Corporation at the time of study conduct. P.C. Arratia reports personal fees from Novartis during the conduct of the study and outside the submitted work. F. Branle reports other support from Novartis AG outside the submitted work. M. Shi reports other support from Novartis during the conduct of the study. D.-W. Kim reports grants and nonfinancial support from Novartis during the conduct of the study as well as grants from Alpha Biopharma, AstraZeneca, Boehringer-Ingelheim, Hanmi, Janssen, Mers, Mirati Therapeutics, Novartis, Ono Pharmaceutical, Pfizer, Roche/Genentech, Takeda, TP Therapeutics, Xcovery, Yuhan, Chong Keun Dang, BioTherapeutics, GSK, and MSD, and grants and personal fees from Amgen and Daiichi-Sankyo outside the submitted work. No disclosures were reported by the other authors.

L.Q.M. Chow: Conceptualization, data curation, formal analysis, validation, writing–review and editing. F. Barlesi: Conceptualization, data curation, validation, writing–review and editing. E.M. Bertino: Data curation, validation, writing–review and editing. M.J. van den Bent: Data curation, validation, writing–review and editing. H.A. Wakelee: Conceptualization, data curation, validation, writing–review and editing. P.Y. Wen: Conceptualization, data curation, validation, writing–review and editing. C.-H. Chiu: Data curation, validation, writing–review and editing. S. Orlov: Data curation, formal analysis, validation, writing–review and editing. R. Chiari: Data curation, formal analysis, validation, writing–review and editing. M. Majem: Data curation, validation, writing–review and editing. M. McKeage: Data curation, validation, writing–review and editing. C.-J. Yu: Data curation, validation, writing–review and editing. P. Garrido: Data curation, validation, writing–review and editing. F.K. Hurtado: Data curation, validation, writing–review and editing. P.C. Arratia: Conceptualization, data curation, formal analysis, validation, writing–review and editing. Y. Song: Conceptualization, data curation, formal analysis, validation, writing–review and editing. F. Branle: Conceptualization, validation, writing–review and editing. M. Shi: Data curation, formal analysis, validation, writing–review and editing. D.-W. Kim: Conceptualization, data curation, formal analysis, validation, writing–review and editing.

This study was funded by Novartis Pharmaceuticals Corporation. The authors thank the patients, their families and caregivers, and the investigators who participated in the study. Medical writing and editorial assistance was provided by Aarti Kamaraj and Manoj Kumar Patel (Novartis Healthcare Pvt. Ltd.).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.