14 CLINICAL STUDIES

The efficacy of COSELA was established in three randomized, double-blind, placebo-controlled trials in patients with extensive stage-small cell lung cancer (ES-SCLC). Study 1 (G1T28-05; NCT03041311) enrolled adult patients receiving carboplatin, etoposide, and atezolizumab for newly diagnosed ES-SCLC. Study 2 (G1T28-02; NCT02499770) enrolled adult patients receiving etoposide/carboplatin for newly diagnosed ES-SCLC. Study 3 (G1T28-03; NCT02514447) enrolled adult patients receiving topotecan for previously treated ES-SCLC.

During Cycle 1, all three studies prohibited primary prophylactic granulocyte-colony stimulating factor (G-CSF) and erythropoiesis-stimulating agent (ESA) use. Both ESAs and prophylactic G-CSF were allowed from Cycle 2 onwards as clinically indicated. Therapeutic G-CSF, RBC, and platelet transfusions were allowed at any time during the studies as clinically indicated.

Study 1: COSELA Prior to Etoposide, Carboplatin, and Atezolizumab (E/P/A)

Patients with newly diagnosed ES-SCLC not previously treated with chemotherapy

Study 1 (G1T28-05) was a randomized (1:1), double-blind, placebo-controlled study of COSELA or placebo administered prior to treatment with etoposide, carboplatin, and atezolizumab (E/P/A) for patients with newly diagnosed ES-SCLC not previously treated with chemotherapy.

A total of 107 patients were randomized to receive COSELA (n=54) or placebo (n=53) prior to administration of E/P/A; patients were stratified by Eastern Cooperative Oncology Group (ECOG) performance status (0 to 1 vs 2) and the presence of brain metastases. Carboplatin (AUC 5) and atezolizumab (1200 mg) were administered on Day 1 and etoposide (100 mg/m2) and COSELA (240 mg/m2) or placebo were administered on Days 1, 2, and 3 of a 21-day cycle for a maximum of 4 cycles (induction). After induction, maintenance atezolizumab (1200 mg) monotherapy on Day 1 of a 21-day cycle continued until disease progression or unacceptable toxicity. COSELA was not administered during maintenance.

The study population characteristics were: median age 64 years (range: 45 to 83); 70% male; 97% white; 14% ECOG performance status 2; 28% with a history of brain metastases; 38% current smokers; 46% lactate dehydrogenase (LDH) >ULN.

The mean relative dose intensities (RDIs) for E/P/A in patients receiving COSELA were 93%, 95%, and 93%, respectively. The mean RDIs for E/P/A in patients receiving placebo were 88%, 89%, and 91%, respectively. Dose reductions of carboplatin occurred in 2% of patients receiving COSELA and in 25% of patients receiving placebo; dose reductions of etoposide occurred in 6% of patients receiving COSELA and in 26% of patients receiving placebo. No dose reduction was allowed for COSELA or atezolizumab.

The study demonstrated a statistically significantly shorter duration of severe neutropenia (DSN) in Cycle 1 (0 vs 4 days) and a lower proportion of patients with severe neutropenia (SN) (2% vs 49%) in patients receiving COSELA compared with placebo (Table 5). Nineteen percent of patients receiving COSELA had Grade 3 or 4 decreased hemoglobin compared with 28% of patients receiving placebo (adjusted relative risk 0.663 [95% CI: 0.336, 1.310]). The rate of RBC transfusions over time was 1.7/100 weeks for patients receiving COSELA and 2.6/100 weeks for patients receiving placebo (adjusted relative risk was not estimable). Six percent of patients receiving COSELA received erythropoiesis-stimulating agents (ESAs) compared with 11% of patients receiving placebo (adjusted relative risk 0.529 [95% CI: 0.145, 1.927]).

Table 5: Study 1: Myeloprotective Efficacy Results in Patients Treated with COSELA or Placebo Prior to Chemotherapy (Intent-to-Treat Analysis)
Endpoint	COSELA 240 mg/m2 (N=54)	Placebo (N=53)	Treatment Effecta (Mean Difference* or Adjusted Relative Risk) (95% CI)	Adjusted 1-sided p-valueb
ANCOVA=analysis of covariance; CI=confidence interval; DSN=duration of severe neutropenia; G-CSF=granulocyte colony-stimulating factor; N=total number of patients in each treatment group; RBC=red blood cell; SD=standard deviation
a The following statistical models were used to assess treatment effects: non-parametric ANCOVA (DSN in Cycle 1); modified Poisson regression (occurrence of SN and RBC transfusion on/after 5 weeks); negative binomial regression (number of all-cause dose reductions). All models included the following as covariates: ECOG status, presence of brain metastases, and the corresponding baseline laboratory values.
b One-sided adjusted p-value obtained from a Hochberg-based gatekeeping procedure.
Primary Endpoints
DSN in Cycle 1 (days): mean (SD)	0 (1.0)	4 (4.7)	-3.6* (-4.9, -2.3)	<0.0001
Number (%) of patients with severe neutropenia	1 (1.9%)	26 (49.1%)	0.038 (0.008, 0.195)	<0.0001
Key Secondary Endpoints
Number of all-cause dose reductions, event rate per cycle	0.021	0.085	0.242 (0.079, 0.742)	0.0195
Number (%) of patients with RBC transfusion on/after 5 weeks	7 (13.0%)	11 (20.8%)	0.642 (0.294, 1.404)	--
Number (%) of patients with G-CSF administration	16 (29.6%)	25 (47.2%)	0.646 (0.403, 1.034)	--

Study 2: COSELA Prior to Etoposide and Carboplatin

Patients with newly diagnosed ES-SCLC not previously treated with chemotherapy

Study 2 (G1T28-02) was a randomized (1:1), double-blind, placebo-controlled evaluation of COSELA or placebo administered prior to treatment with etoposide and carboplatin (E/P) for patients with newly diagnosed ES-SCLC not previously treated with chemotherapy. A total of 77 patients were randomized to COSELA (n=39) or placebo (n=38) and stratified by ECOG performance status (0 to 1 vs 2). Carboplatin (AUC 5) was administered on Day 1 and etoposide (100 mg/m2) and COSELA (240 mg/m2) or placebo were administered on Days 1, 2, and 3 of a 21-day cycle until disease progression or unacceptable toxicity.

Ten percent of patients receiving COSELA had Grade 3 or 4 decreased hemoglobin compared with 18% of patients receiving placebo. The rate of RBC transfusions over time was 0.5/100 weeks for patients receiving COSELA and 1.9/100 weeks for patients receiving placebo. Three percent of patients receiving COSELA received ESAs compared with 5% of patients receiving placebo.

Table 6: Study 2: Myeloprotective Efficacy Results in Patients Treated with COSELA or Placebo Prior to Chemotherapy (Intent-to-Treat Analysis)
Endpoint	COSELA 240 mg/m2 (N=39)	Placebo (N=38)
DSN=duration of severe neutropenia; G-CSF=granulocyte colony-stimulating factor; N=total number of patients in each treatment group; RBC=red blood cell; SD=standard deviation
DSN in Cycle 1 (days): mean (SD)	0 (0.5)	3 (3.9)
Number (%) of patients with severe neutropenia	2 (5.1%)	16 (42.1%)
Number of all-cause dose reductions, event rate per cycle	0.022	0.084
Number (%) of patients with RBC transfusion on/after 5 weeks	2 (5.1%)	9 (23.7%)
Number (%) of patients with G-CSF administration	4 (10.3%)	24 (63.2%)

Study 3: COSELA Prior to Topotecan

Patients with ES-SCLC previously treated with chemotherapy

Study 3 (G1T28-03) included a randomized, double-blind, placebo-controlled evaluation of COSELA or placebo administered prior to topotecan in patients with ES-SCLC previously treated with chemotherapy. A total of 61 patients were randomized to COSELA (n=32) or placebo (n=29). Patients were stratified by ECOG performance status (0 to 1 vs 2) and sensitivity to first-line treatment. Topotecan (1.5 mg/m2) and COSELA (240 mg/m2) or placebo were administered on Days 1-5 of a 21-day cycle. Treatment was administered until disease progression or unacceptable toxicity.

Thirty-eight percent of patients receiving COSELA had Grade 3 or 4 decreased hemoglobin compared with 59% of patients receiving placebo. The rate of RBC transfusions over time was 2.6/100 weeks for patients receiving COSELA and 6.3/100 weeks for patients receiving placebo. Three percent of patients receiving COSELA received ESAs compared with 21% of patients receiving placebo.

Table 7: Study 3: Myeloprotective Efficacy Results in Patients Treated with COSELA or Placebo Prior to Chemotherapy (Intent-to-Treat Analysis)
Endpoint	COSELA 240 mg/m2 (N=32)	Placebo (N=29)
DSN=duration of severe neutropenia; G-CSF=granulocyte colony-stimulating factor; N=total number of patients in each treatment group; RBC=red blood cell; SD=standard deviation
Primary Endpoints
DSN in Cycle 1 (days): mean (SD)	2 (3.9)	7 (6.2)
Number (%) of patients with severe neutropenia	13 (40.6%)	22 (75.9%)
Key Secondary Endpoints
Number of all-cause dose reductions, event rate per cycle	0.051	0.116
Number (%) of patients with RBC transfusion on/after 5 weeks	10 (31.3%)	12 (41.4%)
Number (%) of patients with G-CSF administration	16 (50.0%)	19 (65.5%)
Number (%) of patients with platelet transfusion	8 (25.0%)	9 (31.0%)

Patient Subgroups

Treatment efficacy was examined across different subgroups, including those quantifying the risk of febrile neutropenia, anemia, and RBC transfusions; results were consistent across subgroups.