Tarlatamab, a First-in-Class DLL3-Targeted Bispecific T-Cell Engager, in Recurrent Small-Cell Lung Cancer: An Open-Label, Phase I Study

PURPOSE

Small-cell lung cancer (SCLC) is an aggressive malignancy with limited treatments. Delta-like ligand 3 (DLL3) is aberrantly expressed in most SCLC. Tarlatamab (AMG 757), a bispecific T-cell engager molecule, binds both DLL3 and CD3 leading to T-cellb–mediated tumor lysis. Herein, we report phase I results of tarlatamab in patients with SCLC.

PATIENTS AND METHODS

This study evaluated tarlatamab in patients with relapsed/refractory SCLC. The primary end point was safety. Secondary end points included antitumor activity by modified RECIST 1.1, overall survival, and pharmacokinetics.

RESULTS

By July 19, 2022, 107 patients received tarlatamab in dose exploration (0.003 to 100 mg; n = 73) and expansion (100 mg; n = 34) cohorts. Median prior lines of anticancer therapy were 2 (range, 1-6); 49.5% received antiprogrammed death-1/programmed death ligand-1 therapy. Any-grade treatment-related adverse events occurred in 97 patients (90.7%) and grade b % 3 in 33 patients (30.8%). One patient (1%) had grade 5 pneumonitis. Cytokine release syndrome was the most common treatment-related adverse event, occurring in 56 patients (52%) including grade 3 in one patient (1%). Maximum tolerated dose was not reached. Objective response rate was 23.4% (95% CI, 15.7 to 32.5) including two complete and 23 partial responses. The median duration of response was 12.3 months (95% CI, 6.6 to 14.9). The disease control rate was 51.4% (95% CI, 41.5 to 61.2). The median progression-free survival and overall survival were 3.7 months (95% CI, 2.1 to 5.4) and 13.2 months (95% CI, 10.5 to not reached), respectively. Exploratory analysis suggests that selecting for increased DLL3 expression can result in increased clinical benefit.

CONCLUSION

In patients with heavily pretreated SCLC, tarlatamab demonstrated manageable safety with encouraging response durability. Further evaluation of this promising molecule is ongoing.

INTRODUCTION

Small-cell lung cancer (SCLC) is an aggressive lung cancer subtype with neuroendocrine differentiation diagnosed in more than 150,000 people worldwide each year.^1,2 The 3-year survival rate for patients with extensive stage (ES) SCLC is 6%.³ The addition of immune checkpoint inhibitors atezolizumab or durvalumab to platinum and etoposide chemotherapy followed by maintenance therapy with checkpoint inhibitor alone as first-line treatment for SCLC has led to approximately 30% reduction in the risk of death and durable but modest survival gains for a small subset of patients with ES-SCLC.^3,4 Available therapies are limited for the majority of patients with SCLC who relapse. Topotecan, the most widely used second-line agent globally, has limited efficacy and an unfavorable safety profile.^5,6 Lurbinectedin, in 2020 became the first drug approval by the US Food and Drug Administration in over 20 years for second line, was conditionally approved on the basis of an objective response rate (ORR) of 35%; however, a randomized study failed to demonstrate OS benefit.^7,8 No agent is specifically approved for third-line treatment of relapsed SCLC.

CONTEXT

• Key Objective

• Tarlatamab, a half-life extended bispecific T-cell engager molecule targeting delta-like ligand 3, is the first agent in this new therapeutic class to be evaluated clinically in patients with small-cell lung cancer (SCLC). An analysis of data from the large phase I study of tarlatamab in patients with relapsed/refractory SCLC was performed to assess safety and efficacy.

• Knowledge Generated

• With more than 100 patients treated in the phase I study, tarlatamab has shown tolerability and remarkable antitumor activity over a wide range of target doses in patients with heavily pretreated SCLC. A promising response rate, high median duration of response, and encouraging median overall survival, along with a reasonable safety profile, observed in this study form the basis for planned and ongoing later-stage studies.

• Relevance (T.E. Stinchcombe)

• Tarlatamab has a novel mechanism of action, and the activity in previously treated SCLC, a patient population with few therapeutic options, is promising. The results of additional studies are eagerly awaited.*

  *Relevance section written by JCO Associate Editor Thomas E. Stinchcombe, MD.

The notch signaling pathway is a regulator of neuroendocrine differentiation in SCLC.^9,10 The inhibitory notch ligand delta-like ligand 3 (DLL3) is aberrantly expressed on the surface of up to 85% of SCLC cells and minimally expressed in normal tissues, making it a compelling therapeutic target.^11-14 In vitro SCLC models have indicated a role for DLL3 in promoting tumor growth, migration, and invasion.¹⁵ The DLL3-targeted antibody-drug conjugate rovalpituzumab tesirine showed clinical antitumor activity in patients with SCLC.¹⁶

Tarlatamab, a half-life extended bispecific T-cell engager (HLE BiTE) molecule, binds both DLL3 on cancer cells and CD3 on T cells leading to T-cell–mediated tumor lysis. Tarlatamab promotes tumor regression in preclinical models of SCLC.¹⁷ Tarlatamab is the first DLL3-targeted immune therapy to be evaluated clinically in SCLC.

In this phase I trial, we evaluated the safety, pharmacokinetics (PK), and preliminary efficacy of tarlatamab in patients with SCLC.

PATIENTS AND METHODS

Study Design and Participants

DeLLphi-300 (ClinicalTrials.gov identifier: NCT03319940) is a phase I, multicountry, open-label, dose-escalation study evaluating tarlatamab monotherapy and combination with antiprogrammed cell death protein one (PD-1) therapy in patients with SCLC. The results reported here are limited to the monotherapy regimen including dose escalation and expansion cohorts. Eligible patients were age 18 years or older with histologically or cytologically confirmed SCLC who had progressed or recurred after at least one previous platinum-based regimen and, if standard of care, a programmed cell death ligand 1 (PD-L1) inhibitor in addition to chemotherapy. Patients had an Eastern Cooperative Oncology Group performance status of two or less and at least two measurable lesions defined per modified response evaluation criteria in solid tumors (RECIST), version 1.1 and if present, brain metastases that were clinically and radiologically stable after treatment. Key exclusion criteria were untreated brain metastases and severe or recurrent immune-mediated adverse events (AEs) or infusion-related reactions while on prior immunotherapy. Full eligibility criteria are included in the study Protocol (online only).

The protocol and amendments were approved by the institutional review board or ethics committee at each participating site. The trial was conducted in accordance with the International Council for Harmonisation Good Clinical Practice guidelines and the principles of the Declaration of Helsinki. All patients provided written informed consent.

Procedures

The planned tarlatamab dose levels were 0.003, 0.01, 0.03, 0.1, 0.3, 1, 3, 10, 30, and 100 mg by intravenous (IV) infusion administered every 2 weeks. The first four dose levels were planned as single-patient cohorts as AEs were expected to be low. Tarlatamab administration was continued until disease progression, unacceptable side effects, or consent withdrawal.

Outcomes

The primary end point was safety including dose-limiting toxicities (DLTs; defined as tarlatamab-related toxicity within 28 days after the first dose and meeting protocol criteria), AEs during the treatment period (treatment-emergent AEs [TEAEs]), and TEAEs possibly related to tarlatamab per investigator review (treatment-related AEs). Secondary end points included PK, antitumor activity including objective response per modified RECIST 1.1^18-20 by investigator assessment, duration of response (DOR), time to response, progression-free survival (PFS), and overall survival (OS). Exploratory end points assessed included antitarlatamab antibody formation and relationship between baseline target protein expression and clinical benefit (Data Supplement [online only]).

Statistical Analysis

This analysis included patients enrolled in escalation and expansion cohorts. Data cutoff was July 19, 2022. Sample size was based on clinical considerations and standard dose-escalation design. A two-parameter Bayesian Logistic Regression Model guided dose exploration.²¹ The Sponsor, in consultation with investigators, reviewed the Bayesian Logistic Regression Model recommended dose level and cumulative data by cohort before dose escalation decisions. AEs and DLTs were evaluated continually. On the basis of the overall benefit-risk profile of 100 mg, this dose was further evaluated as the expansion dose. Descriptive statistics are provided for selected demographics, safety, PK, pharmacodynamics, and biomarker data. Exploratory analysis was performed to evaluate the relationship between baseline expression of DLL3 and clinical benefit (Data Supplement). Kaplan-Meier methods were used to estimate the median and percentiles for time to event end points with 95% CI calculated using the Brookmeyer and Crowley²² method.

RESULTS

Patients

Between December 26, 2017, and April 28, 2022, 107 patients received tarlatamab in dose escalation (0.003-100 mg; n = 73) and expansion (100 mg; n = 34) cohorts (Data Supplement). Step dosing was used starting with the 3-mg cohort (using 1 mg as the run-in dose, followed by target dose on day 8, day 15, and every 2 weeks thereafter) because of observed cytokine release syndrome (CRS) in prior cohorts.

Baseline characteristics are summarized in Table 1 (additional in Data Supplement). The median age was 63 years (range, 32-80). Eastern Cooperative Oncology Group performance status was 0-1 in 99% of patients. More than 70% of patients had b % 2 lines of prior therapy, 25% were platinum refractory, and 50% had prior PD-1/PD-L1 inhibitor.

TABLE 1.

Patient Demographics and Baseline Characteristics

The median follow-up was 8.7 months (range, 0.2-31.8). Treatment was discontinued in 92 patients (86%), most commonly for disease progression (n = 77 [72%]). At data cutoff, 47 patients (43.9%) had ended study because of death. The median number of treatment cycles started was 3 (interquartile range [IQR], 1-8), and the median number of tarlatamab doses received was 6 (IQR, 3-16).

Retrospective DLL3 immunohistochemistry analysis showed that DLL3 was expressed (b % 1%) in 85 of 90 (94%) evaluable patients; the median H-score was 186 (range, 0-300), and the median tumor cell positivity was 95% (range, 0%-100%).

Safety and Tolerability

DLTs occurred in six patients including pneumonitis (n = 1 [last prior dose, 0.3 mg]), increased alanine aminotransferase (n = 1 [1 mg]), CRS (n = 1 [1 mg]), encephalopathy (n = 1 [10 mg]), chills, pyrexia, and neutropenia (n = 1 each [100 mg]). A maximum tolerated dose was not reached; the highest protocol-planned dose (100 mg) was evaluated in the expansion cohort. Four patients (3.7%) discontinued tarlatamab because of treatment-related AEs of encephalopathy (n = 1), immune effector cell-associated neurotoxicity (ICANS; n = 1), and pneumonitis (n = 2). A single G5 pneumonitis event was recorded in a 70-year old man with a history of prior carboplatin/etoposide chemotherapy, chronic obstructive pulmonary disease, and radiation to the lung and pleural nodules. The event onset was cycle 1 day 18, 3 days after the second tarlatamab treatment (both doses 0.3 mg), and was confounded by clinically significant disease progression at the time of pneumonitis requiring urgent palliative radiation to the lung and to a soft tissue mass in the thoracic spine causing spinal cord compression. The cause of death was attributed by the investigator to disease progression and pneumonitis. An additional G3 and three additional G2 TEAEs of pneumonitis were observed (5 of 107 [4.7%] overall incidence of pneumonitis). Among patients with G2 pneumonitis, one patient ended treatment because of neurotoxicity (not pneumonitis), one patient had resolution of pneumonitis before discontinuation for PD, and one patient resumed treatment without dose change.

TEAEs of any cause/grade occurred in 107 patients (100%; Table 2 and Data Supplement). The most common were CRS (56 patients [52.3%]), pyrexia (43 [40.2%]), and constipation (33 [30.8%]). Grade b % 3 AEs occurred in 61 patients (57.0%) with the most common being neutropenia (8.4%; Data Supplement). Serious AEs occurred in 55 patients (51.4%; Data Supplement). TEAEs led to dose reductions in nine patients (8.4%) with 4 (3.7%) having CRS-related reductions (Data Supplement). Dose interruption occurred in 20 patients (18.7%; Data Supplement). Any-grade and grade b % 3 treatment-related AEs occurred in 97 (90.7%) and 33 (30.8%) patients, respectively (Data Supplement).

TABLE 2.

AEs (preferred term and AMQ for selected terms)

CRS, neutropenia, and neurologic events were monitored as events of interest on the basis of preclinical, clinical, and mechanistic data with tarlatamab, other BiTE molecules, and other T-cell–associated therapies. Amgen MedDRA Query narrow searches were performed to supplement standard system organ class single preferred term safety reporting (defined in Data Supplement and summarized in Table 2). Measures to ameliorate the potential for CRS included prophylactic corticosteroids (cycle 1 only) and IV hydration. Grade b % 2 treatment-emergent CRS was reported in 15 patients (14.0%) and grade 3 CRS in one patient (0.9%); no grade 4 or 5 CRS has been reported. In patients where time of onset of CRS as well as date of onset of CRS was recorded (n = 47), the median time to onset of CRS after the last dose was 17.5 hours. CRS was transient (median duration, 3 days [IQR, 2-4 days]) and resolved in all cases. Eight patients (7.5%) received tocilizumab for CRS. CRS was largely confined to cycle 1. A total of five patients (4.7%) had CRS in cycle 2 or later; four had prior CRS in cycle 1 while one first experienced CRS in cycle 2. Treatment-emergent neurologic AEs of any grade occurred in 75 patients (70.1%) and were mostly grade 1: dysgeusia (29.0%), headache (19.6%), and dizziness (10.3%) were most common. Grade b % 3 treatment-emergent neurologic events occurred in 12 patients (11.2%) including confusional state (4.7%), delirium (1.9%), and encephalopathy (1.9%). One patient had a grade 4 neurologic event (confusion), and none had grade 5. All grade b % 3 neurologic AEs resolved, with one patient discontinuing tarlatamab because of G3 encephalopathy and two other patients continuing treatment at reduced doses. G2 ICANS was the other neurologic cause leading to discontinuation in one patient. First onset of any grade neurological event was mostly within the first 30 days of treatment (median, 9 days [IQR, 2-29 days]) with a median duration of 5 days (IQR, 2-15 days). Grade b % 3 neutropenia occurred in 11 patients (10.3%). Any-grade neutropenia first onset occurred at a median of 30 days (IQR, 21-31 days) after first tarlatamab administration, and the median duration was 7 days (IQR, 4-13); overall, 10 patients (9.3%) received G-CSF. Febrile neutropenia occurred in one patient and was not considered treatment related.

Efficacy

Confirmed ORR was 23.4% (95% CI, 15.7 to 32.5) including two complete and 23 partial responses by modified RECIST 1.1 per investigator assessment (Table 3). Figure 1A illustrates best percentage change from baseline in sum of diameters for patients with evaluable postbaseline assessments (n = 94). The disease control rate was 51.4% (95% CI, 41.5 to 61.2). Responses were seen starting with the 0.3 mg dose and generally higher rates of response were observed at doses of 3 mg and above. At least 30% tumor shrinkage in target lesions at postbaseline assessment was observed in 39 patients (36.4%). Among confirmed responders, the median time to response was 1.8 months (range, 1.2-7.4) and the median DOR was 12.3 months (95% CI, 6.6 to 14.9; Fig 1B). The longest DOR was 14.9 months, and 11 patients (44% of responders) had ongoing response at data cutoff. The median PFS was 3.7 months (95% CI, 2.1 to 5.4), and the median OS was 13.2 months (95% CI, 10.5 to NE; Fig 2), respectively. Of 77 progressive disease events, eight were in the brain (10.4%). Twenty-eight patients (26.2%) received subsequent anticancer therapies after tarlatamab. Representative computed tomography scans from patients receiving tarlatamab are provided (Data Supplement). An exploratory post hoc analysis demonstrates the sensitivity and specificity of enriching for responders (using confirmed OR) when total DLL3 expression is retrospectively considered for patient selection (Fig 3). Clinical benefit is observed across a range of thresholds.

TABLE 3.

Tumor Response to Tarlatamab According to Investigator Assessment

FIG 1.

Response to tarlatamab. (A) Best percent change from baseline in tumor burden (defined by the sum of the longest diameters of all target lesions) in 94 patients whose data cutoff date is at least 9 weeks after the first dose date and for whom postbaseline tumor data were available. ^aSD, patients had an initial response but did not have confirmation of response on the subsequent scan and ^bPR, patients had an initial PR and still have potential for future confirmative scans. One confirmed patient in the 100 mg expansion cohort had missing sum of diameters for lesion measurement and was not included in the plot. ^cStep dosing (ie, 1 mg run-in dose) was used in these cohorts. (B) TTR, the duration of treatment, and patient status as of the data cutoff date according to dose of tarlatamab for all patients with confirmed response (n = 25). CR, complete response; NE, not evaluable; PR, partial response; SD, stable disease; TTR, time to response.

FIG 2.

Efficacy of tarlatamab in patients with SCLC. (A) Kaplan-Meier curve of PFS for patients whose data cutoff date is at least 9 weeks after the first dose date (N = 107). (B) Kaplan-Meier curve of OS for patients whose data cutoff date is at least 9 weeks after the first dose date (N = 107). OS, overall survival; PFS, progression-free survival; SCLC, small-cell lung cancer.

FIG 3.

ROC curve of objective response enrichment by DLL3 expression. ROC curve reflecting true-positive rate and false-positive rate of enrichment of objective response when FIH patients (1-100 mg cohorts) are retrospectively examined for selection using DLL3 expression assessed with the Ventana SP347 assay. ROC analysis included 77 patients enrolled in the 1-100 mg cohorts for whom pretreatment DLL3 expression readouts were available. DLL3, delta-like ligand 3; FIH, first-in-human; ROC, receiver operating characteristic.

Clinical PK

As of April 15, 2022, preliminary pharmacokinetic data from dose escalation and expansion cohorts were available for 101 patients. Briefly, tarlatamab exhibited approximate dose proportional increase in serum exposures. Approximate steady state in serum tarlatamab exposures were achieved within 4 weeks of every other week target regimen initiation, with minimal accumulation. The mean (B1 standard deviation) terminal elimination half-life estimated at steady state across the evaluated target dose range was approximately 5.7 (B1 2.2) days, which is consistent with the intended half-life extension of the HLE platform relative to non-HLE BiTEs.²³

Immunogenicity

Among patients with available samples, 10 of 97 (10.3%) developed anti-tarlatamab antibodies after tarlatamab administration. Two of 99 patients (2.0%) had preexisting antibodies at baseline. There is no apparent antidrug antibody impact on tarlatamab exposures or the safety profile in these patients.

Pharmacodynamics

The pharmacodynamic response after first dose of tarlatamab infusion was characterized by initial T-cell redistribution (Data Supplement [Figs 3A and 3B, online only]), T-cell activation (Data Supplement [Fig 3C]), and transient IFN-gamma elevation (Data Supplement [Fig 3D]). For step-dose cohorts, pharmacodynamic responses were greatest after initial administration of 1-mg step dose and not exceeded with target dose administration.

DISCUSSION

Tarlatamab demonstrated a manageable safety profile across a wide dose range through the expansion dose of 100 mg and was associated with encouraging response rates in a heavily pretreated population of patients with SCLC. Confirmed responses were durable, and OS seems promising. Across all doses (N = 107), tarlatamab was discontinued in only 4 (3.7%) patients. A maximum tolerated dose was not reached; the highest dose (100 mg) was further evaluated in dose expansion.

CRS is expected based on the mechanism of action of tarlatamab. CRS was the most frequent TEAE observed (56%) but was generally low-grade, transient, and typically occurred in the first cycle. CRS was reversible and generally managed with antipyretics, IV fluids, and steroids; tocilizumab was used to treat CRS in eight of 107 patients (7.5%). Neutropenia was a risk associated with tarlatamab observed in this study and was unexpected based on preclinical data; the mechanism is not understood. The study protocol was updated accordingly for specific monitoring and management. Further evaluation of neutropenia will be relevant to trials of tarlatamab use in combination with other marrow-suppressing therapies. Neurologic evaluation was conducted as part of frequent clinical evaluation to assess study patients for CRS and/or neurologic AEs due to the known association with immune effector cell therapies. Most neurologic AEs were mild and self-limiting without need for treatment discontinuation or dose reduction, though 12 patients (11.2%) had grade b % 3 neurologic AEs. Neurologic AEs led to tarlatamab discontinuation in two patients (encephalopathy, ICANS). Careful evaluation of neurologic AEs is ongoing to better characterize these events and identify risk factors or interventions that might specifically improve management.

Few approved therapies exist for SCLC after first line. A phase II study of lurbinectedin in second-line SCLC found an ORR of 35% and a median DOR of 5.3 months.⁷ In a randomized study of topotecan versus combination chemotherapy in recurrent SCLC, topotecan ORR was 24% and the median DOR was 3.3 months.⁵ The prior conditional approval by US Food and Drug Administration of nivolumab and pembrolizumab for third-line or later SCLC was based on response rates of 12% and 19%, respectively, with durable responses seen at b % 12 months in > 60% of responding patients.²⁴ These approvals were subsequently withdrawn as survival benefit was not demonstrated.²⁵ The ORR of 23% and median DOR of 12.3 months for tarlatamab compares well with other therapies, especially considering that more than 70% of patients had at least two prior lines of therapy. Fifty percent of patients in this study had received prior PD-1/PD-L1 therapy, representative of current practice in first-line SCLC. Despite the relatively short median PFS (3.7 months) seen with tarlatamab, the median OS (13.2 months) is relatively high and compares favorably with 9.3-month median OS reported previously with lurbinectedin or approximately 6-month OS with topotecan, although the value of comparisons is limited by differences in study design and patient populations.^5,7 The promising OS benefit may reflect the long durability of response seen thus far in those who respond to tarlatamab but further follow up is needed in larger randomized studies. Another explanation could be that OS benefit derived from post-tarlatamab treatment, although this is less likely a major factor because only 26.2% of patients received such treatment in this heavily pretreated cohort. Identifying factors predictive of response and/or toxicity (eg, prior therapies) is an ongoing effort. An exploratory analysis suggests that increased DLL3 expression trends with higher magnitude of clinical benefit.

Limitations of this study include moderate median follow-up time (8.7 months) and selected patient population required for a first-in-human trial. Interpretation of preliminary efficacy is limited in a single-arm, dose-ranging study. Efficacy data from several different dose levels was pooled for some analyses; enrollment to the expansion cohort is ongoing.

The first DLL3-targeted agent to enter the clinic was the antibody-drug conjugate rovalpituzumab tesirine (Rova-T).¹¹ During development, data showed strong enrichment in clinical benefit in patients with relatively high DLL3 expression; however, despite implementing DLL3 expression as an eligibility or stratification factor in later phase studies, Rova-T failed to show a survival benefit versus standard of care in two phase III studies and was discontinued.^16,26,27 Rova-T toxicity in preclinical and especially clinical studies turned out to be a significant problem with this construct and was more consistent with general effects of the pyrrolobenzodiazepine payload rather than the DLL3 target.^16,26,28 In contrast, tarlatamab has exhibited tolerability and efficacy over a broad range of doses in a heavily pretreated population confirming the value of this target. Other DLL3-specific therapies undergoing clinical evaluation include the T-cell engaging molecules HPN328 and BI 764532 and the chimeric antigen receptor T-cell therapy AMG 119.^29-32

These results demonstrate promising activity in patients with high unmet medical need and have led to ongoing investigations of tarlatamab as monotherapy in SCLC and other neuroendocrine cancers (phase II study in patients with third-line SCLC [NCT05060016]; phase Ib study in neuroendocrine prostate cancer [NCT04702737]) and in combination in earlier lines of therapy (phase I study of first-line ES-SCLC tarlatamab in combination with carboplatin, etoposide, and PD-L1 inhibitor [NCT05361395]).

APPENDIX

TABLE A1.

Study Sites and Primary Investigators

DATA SHARING STATEMENT

There is a plan to share data. This may include deidentified individual patient data for variables necessary to address the specific research question in an approved data sharing request, also related data dictionaries, study protocol, statistical analysis plan, informed consent form, and/or clinical study report.

Data sharing requests relating to data in this manuscript will be considered after the publication date and (1) this product and indication (or other new use) have been granted marketing authorization in both the United States and Europe, or (2) clinical development discontinues and the data will not be submitted to regulatory authorities. There is no end date for eligibility to submit a data sharing request for these data.

Qualified researchers may submit a request containing the research objectives, the Amgen product(s) and Amgen study/studies in scope, end points/outcomes of interest, statistical analysis plan, data requirements, publication plan, and qualifications of the researcher(s).

In general, Amgen does not grant external requests for individual patient data for the purpose of reevaluating safety and efficacy issues already addressed in the product labeling. A committee of internal advisors review requests. If not approved, requests may be further arbitrated by a Data Sharing Independent Review Panel. Requests that pose a potential conflict of interest or an actual or potential competitive risk may be declined at Amgen's sole discretion and without further arbitration.

Upon approval, information necessary to address the research question will be provided under the terms of a data sharing agreement. This may include anonymized individual patient data and/or available supporting documents, containing fragments of analysis code where provided in analysis specifications.

Further details are available at the following: https://wwwext.amgen.com/science/clinical-trials/clinical-data-transparency-practices/clinical-trial-data-sharing-request.

AUTHOR CONTRIBUTIONS

Conception and design: Nooshin Hashemi Sadraei

Provision of study materials or patients: Luis Paz-Ares, Stephane Champiat, W. Victoria Lai, Hiroki Izumi, Ramaswamy Govindan, Michael Boyer, Horst-Dieter Hummel, Hossein Borghaei, Melissa L. Johnson, Neeltje Steeghs, Fiona Blackhall, Afshin Dowlati, Noemi Reguart, Tatsuya Yoshida, Kai He, Shirish M. Gadgeel, Enriqueta Felip, Taofeek K. Owonikoko

Collection and assembly of data: Luis Paz-Ares, Stephane Champiat, W. Victoria Lai, Hiroki Izumi, Ramaswamy Govindan, Michael Boyer, Horst-Dieter Hummel, Hossein Borghaei, Melissa L. Johnson, Neeltje Steeghs, Fiona Blackhall, Afshin Dowlati, Noemi Reguart, Tatsuya Yoshida, Kai He, Shirish M. Gadgeel, Enriqueta Felip, Yiran Zhang, Taofeek K. Owonikoko

Data analysis and interpretation: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Tarlatamab, a First-in-Class DLL3-Targeted Bispecific T-Cell Engager, in Recurrent Small-Cell Lung Cancer: An Open-Label, Phase I Study

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).