A protocol pre-specified interim overall survival analysis of the randomised phase III GEMSTONE-302 trial: sugemalimab or placebo plus platinum-based chemotherapy for first-line metastatic non-small-cell lung cancer

doi:10.21203/rs.3.rs-2066919/v1

Download PDF

Article

A protocol pre-specified interim overall survival analysis of the randomised phase III GEMSTONE-302 trial: sugemalimab or placebo plus platinum-based chemotherapy for first-line metastatic non-small-cell lung cancer

https://doi.org/10.21203/rs.3.rs-2066919/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 15 Jun, 2023

Read the published version in Nature Cancer →

Version 1

posted

You are reading this latest preprint version

In the primary analysis of the GEMSTONE-302 trial, sugemalimab plus platinum-based chemotherapy conferred significantly longer progression-free survival (PFS) versus placebo in patients with previously untreated metastatic non-small-cell lung cancer (NSCLC) with no known driver oncogene alterations. Here we report data from a protocol pre-specified interim overall survival (OS) analysis of that trial. In this randomised, double-blind, phase III trial (NCT03789604), adult patients with systemic-treatment-naïve stage IV NSCLC and no known driver oncogene alterations were randomised 2:1 to receive sugemalimab (1200 mg, intravenously) or placebo plus platinum-based chemotherapy every 3 weeks for up to four cycles, followed by maintenance therapy with intravenous sugemalimab or placebo plus pemetrexed (non-squamous NSCLC) or sugemalimab or placebo (squamous NSCLC). Placebo-treated patients could cross over to receive sugemalimab monotherapy upon disease progression. As of 22 November 2021 [median follow-up, 25.4 months (sugemalimab) vs 24.9 months (placebo)], median OS was 25.4 months [95% confidence interval (CI) 20.1-not reached] and 16.9 months (95% CI 12.8–20.7) for the sugemalimab and placebo groups, respectively [hazard ratio (HR) 0.65, 95% CI 0.50–0.84; P = 0.0008]. Median investigator-assessed PFS was 9.0 months (95% CI 7.4–10.9) and 4.9 months (95% CI 4.8–5.2), respectively (HR 0.49, 95% CI 0.40–0.61; P < 0.0001). The OS and PFS benefit with sugemalimab was maintained regardless of histology types and programmed death-ligand 1 (PD-L1) expression. Sugemalimab-treated patients had a significantly higher objective response rate (ORR) versus placebo (63.4% versus 40.3%; P < 0.0001) with a longer median duration of response [9.9 months (95% CI 8.6–13.2) versus 4.4 months (95% CI 3.5–6.1)]. There were no new safety signals. Sugemalimab plus chemotherapy conferred a statistically significant and clinically meaningful improvement in OS and PFS regardless of PD-L1 expression and histology types, supporting sugemalimab as a new first-line treatment option for metastatic NSCLC.

sugemalimab

PD-L1

combination therapy

first-line

metastatic NSCLC

overall survival

The combination of standard, platinum-based chemotherapy with the programmed death receptor 1 (PD-1) monoclonal antibody pembrolizumab was evaluated against placebo plus platinum-based chemotherapy in patients with previously untreated metastatic (stage IV) non-small-cell lung cancer (NSCLC) in the phase III KEYNOTE-189 (non-squamous NSCLC) and KEYNOTE-407 trials (squamous NSCLC).^1,2 In both trials, the pembrolizumab group was associated with improvements in overall survival (OS) and progression-free survival (PFS) over the placebo group. In patients with non-squamous NSCLC, rates of OS at 36 months were 31.3% with pembrolizumab and 17.4% with placebo.³ In patients with squamous NSCLC, 24-month OS rates were 37.5% and 30.6%, respectively.² Addition of the programmed death ligand 1 (PD-L1) monoclonal antibody atezolizumab to platinum-based chemotherapy in patients with treatment-naïve, stage IV NSCLC in two separate phase III trials – IMpower132 (non-squamous NSCLC) and IMpower131 (squamous NSCLC) – achieved improvement in PFS relative to the placebo combination, but no OS benefit.^4,5 Moreover, data from the phase III POSEIDON trial showed similar findings for first-line combination therapy with another anti-PD-L1 inhibitor, durvalumab, in patients with metastatic non-squamous or squamous NSCLC. Durvalumab plus platinum-based chemotherapy significantly improved PFS over placebo, with no significant improvement in OS.⁶ The OS benefit with first-line anti-PD-L1 plus chemotherapy in these two pathological types has yet to be established. Several phase III trials of similar combinations in patients with first-line advanced NSCLC in China have also demonstrated PFS benefits over chemotherapy alone; however, protocol-prespecified OS analyses have either not yet been reported for these trials,^7,8 or are still immature.⁹

Sugemalimab (formerly CS1001) is a full-length, fully human immunoglobulin G4 (s228p) monoclonal anti-PD-L1 antibody with anti-tumour activity, either as a monotherapy or in combination with platinum-based chemotherapy, in patients across a variety of advanced lymphomas and solid tumours, including non-squamous and squamous NSCLC.^10,11 A multicentre, phase III trial (GEMSTONE-302; ClinicalTrials.gov ID: NCT03789604) was conducted to evaluate the efficacy and safety of sugemalimab plus platinum-based chemotherapy (sugemalimab group) versus placebo plus platinum-based chemotherapy (placebo group) as a first-line treatment in patients with metastatic non-squamous or squamous NSCLC with no known driver oncogene alterations.¹¹ At the pre-specified interim PFS analysis (median follow-up, 8.6 months), GEMSTONE-302 met its primary end point, showing significantly longer investigator-assessed PFS in the sugemalimab group compared with the placebo group. This improvement was sustained at the final PFS analysis after a median follow-up duration of 17.8 months [median PFS 9.0 months (95% CI 7.4–10.8) versus 4.9 months (95% CI 4.8–5.1); stratified hazard ratio (HR) 0.48, 95% confidence interval (CI) 0.39–0.60; P < 0.0001].¹¹ This benefit appeared to be independent of tumour PD-L1 expression for both non-squamous and squamous NSCLC.

Based on the PFS findings, sugemalimab combined with chemotherapy has received approval from China’s National Medical Products Administration for the first-line treatment of patients with metastatic non-squamous NSCLC with no epidermal growth factor receptor (EGFR) or anaplastic lymphoma kinase (ALK) tumour aberrations, and for patients with metastatic squamous NSCLC.^12,13

Here we report efficacy and safety outcomes from the protocol-specified interim analysis of OS for the GEMSTONE-302 study. We also present, for the first time, the outcomes for patients who crossed over from the placebo group to the sugemalimab group.

Patients and treatments

Of the 846 patients screened between 13 December 2018 and 15 May 2020, 320 were ultimately assigned to the sugemalimab group and 159 to the placebo group. The patient demographics and baseline disease characteristics were similar across the treatment groups (Table S3). All randomised patients were included in the efficacy and safety analysis sets (Fig. 1). As of 22 November 2021, the median duration of follow-up was 25.4 months [interquartile range (IQR) 23.0-28.6] for the sugemalimab group and 24.9 months (IQR 22.4–28.1) for the placebo group. The median duration of study treatment in the double-blind phase was 7.2 months (range 0.2–34.7) and 4.6 months (range 0.3–31.8) for the sugemalimab and placebo groups, respectively; 257 (80.3%) and 124 (78.0%) patients, respectively, had completed four cycles of chemotherapy (Table S4). The median duration of treatment in patients who completed four cycles of platinum-based chemotherapy was 9.8 months (range 3.4–34.7) for the sugemalimab group and 5.2 months (range 3.3–31.8) for the placebo group. A total of 269 (84.1%) patients in the sugemalimab group and 152 (95.6%) in the placebo group had discontinued treatment, primarily due to radiographic evidence of disease progression [184 (57.5%) and 118 (74.2%)]. A total of 174 (54.4%) patients in the sugemalimab group and 113 (71.1%) in the placebo group had discontinued from the study, mainly because of death (Fig. 1). Subsequent anti-cancer therapy was received by 157 (49.1%) and 104 (65.4%) patients in the sugemalimab and placebo groups, respectively. Non-study anti-PD-1/PD-L1 monoclonal antibodies were received by 41 (12.8%) patients in the sugemalimab group and 29 (18.2%) patients in the placebo group; bevacizumab by 29 (9.1%) and 14 (8.8%), respectively; and taxanes by 88 (27.5%) and 49 (30.8%), respectively (Fig. 1). Upon experiencing disease progression, 18 (5.6%) patients from the double-blind sugemalimab group received subsequent sugemalimab monotherapy, and 45 (28.3%) from the placebo group crossed over to receive sugemalimab. A summary of subsequent anti-cancer therapies, including by pathological type, is provided in Table S5.

Overall survival

As of data cut-off date, 253 patients had died [156 (49%) of 320 patients in the sugemalimab group and 97 (61%) of 159 in the placebo group]. Median OS was significantly longer in the sugemalimab group than that in the placebo group [25.4 months (95% CI 20.1-not reached) vs 16.9 months (95% CI 12.8–20.7); HR 0.65, 95% CI 0.50–0.84; P = 0.0008]. Survival rates were 51.7% vs 35.6%, respectively, at 24 months and 45.7% vs 27.7%, respectively, at 30 months (Fig. 2A). Analyses for OS adjusted for patients in the placebo group receiving any subsequent anti-PD-1/PD-L1 immune checkpoint inhibitor therapy, including the protocol-specified treatment crossover, yielded HRs for sugemalimab plus chemotherapy versus placebo plus chemotherapy of 0.58 (95% CI 0.43–0.80), 0.40 (95% CI 0.31–0.53), and 0.55 (95% CI 0.42–0.71) based on the IPCW, RPSFTM, and two-stage AFTM, respectively (Table S6). Analyses for OS adjusted for protocol-specified treatment crossover yielded HRs (sugemalimab combination versus chemotherapy combination) of 0.59 (95% CI 0.45–0.77), 0.68 (95% CI 0.48–0.99), and 0.49 (95% CI 0.31–0.73) in the overall population, and non-squamous and squamous subgroups, respectively (Table S7). Pre-specified subgroup analyses suggested that sugemalimab conferred an OS benefit over placebo in all analysed subgroups, including in patients with non-squamous and squamous NSCLC, and those with tumour PD-L1 ≥ 1% or PD-L1 < 1% (Figs. 2B, 3, and 4). The subgroup analysis also suggested a survival benefit for patients with brain metastases, with a median OS of 22.1 months and 9.0 months in the sugemalimab and placebo groups, respectively (HR 0.45, 95% CI 0.23–0.88) (Fig. 2B).

Progression free survival

At the time of the data cut-off, 246 patients (77%) in the sugemalimab group and 141 patients (89%) in the placebo group had investigator-assessed PFS events. In the intention-to-treat population, investigator-assessed PFS was 9.0 months (95% CI 7.4–10.9) in the sugemalimab group and 4.9 months (95% CI 4.8–5.2) in the placebo group (P < 0.0001). PFS was significantly longer with sugemalimab than without for patients with tumour PD-L1 ≥ 1% at 10.9 months (95% CI 8.9–11.8) versus 4.9 months (95% CI 4.7–5.9), with HR 0.48 (95% CI 0.36–0.6; P < 0.0001). Patients with tumour PD-L1 < 1% also experienced a PFS benefit with sugemalimab compared with placebo (HR 0.57, 95% CI 0.41–0.78) (Table 1).

Table 1

Investigator-assessed ORR and duration of response in patients who received sugemalimab or placebo plus platinum-based chemotherapy as a first-line therapy in the double-blind phase
	Sugemalimab plus chemotherapy	Placebo plus chemotherapy
Objective response rate (CR + PR), n/N (%)	203/320 (63.4)	64/159 (40.3)
95% CI^a	57.9–68.7	32.6–48.3
Difference, % (95% CI)^b	23.2 (13.9–32.5)
P value	< 0.0001
Objective response rate by pathological type, n/N (%)
Non-squamous	112/191 (58.6)	35/96 (36.5)
Squamous	91/129 (70.5)	29/63 (46.0)
Objective response rate by PD-L1 expression, n/N (%)
< 1%	64/124 (51.6)	25/64 (39.1)
≥ 1%	139/196 (70.9)	39/95 (41.1)
Duration of response^c, median (95% CI^d), months	9.9 (8.6–13.2)	4.4 (3.5–6.1)
Duration of response, median (95% CI) by pathological type, months
Non-squamous	10.4 (8.9–15.2)	6.2 (4.0-7.8)
Squamous	9.7 (6.1–13.4)	3.5 (3.3–4.4)
Duration of response, median (95% CI) by PD-L1 expression, months
PD-L1 < 1%	7.5 (5.6–11.7)	3.7 (2.9–6.3)
PD-L1 ≥ 1%	11.5 (9.6–16.6)	5.1 (3.5–6.7)
PFS, median (95% CI), months	9.0 (7.4–10.9)	4.9 (4.8–5.2)
HR, 95% CI	0.49 (0.40–0.61)
P value	< 0.0001
PFS, median (95% CI) by PD-L1 expression, months
PD-L1 < 1%	7.4 (6.8-9.0)	4.9 (4.0-5.8)
HR, 95% CI	0.57 (0.41–0.78)
P value	0.0005
PD-L1 ≥ 1%	10.9 (8.9–11.8)	4.9 (4.7–5.9)
HR, 95% CI	0.48 (0.36–0.63)
P value	< 0.0001
CI, confidence interval; CR, complete response; ECOG PS, Eastern Cooperative Oncology Group performance status; HR, hazard ratio; PD-L1, programmed death ligand 1; PFS, progression-free survival; PR, partial response. ^aCalculated using the Clopper-Pearson method. ^bCochran-Mantel-Haenszel test stratified by ECOG PS (0 versus 1), tumour pathological type (non-squamous versus squamous), and tumour PD-L1 expression (< 1% versus ≥ 1%). ^cKaplan–Meier estimates. ^dComputed using the method of Brookmeyer and Crowley.

Table 2. Summary of AEs, and sugemalimab- or placebo-related AEs by highest NCI-CTCAE grade occurring in ≥ 10% of patients in the sugemalimab or placebo groups (safety analysis set)
	Sugemalimab plus chemotherapy (n = 320)			Placebo plus chemotherapy (n = 159)
Any AE	319 (99.7)			157 (98.7)
Sugemalimab or placebo related	264 (82.5)			103 (64.8)
Grade ≥ 3	207 (64.7)			99 (62.3%)
Sugemalimab or placebo related	95 (29.7)			39 (24.5)
Serious AEs	110 (34.4)			49 (30.8)
Sugemalimab or placebo related	53 (16.6)			16 (10.1)
AEs of special interest	77 (24.1)			5 (3.1)
Grade ≥ 3	10 (3.1)			0
Leading to death	20 (6.3)			9 (5.7)
Any drug related	11 (3.4)			2 (1.3)
Leading to treatment discontinuation (sugemalimab or placebo)	45 (14.1)			12 (7.5)
Leading to treatment interruption (sugemalimab or placebo)	2 (0.6)			0
Leading to treatment cycle delay (sugemalimab or placebo)	144 (45.0)			59 (37.1)
AEs related to either sugemalimab or placebo (≥ 10% in any group) by preferred term	Any grade n (%)	Grade 1-2 n (%)	Grade ≥ 3 n (%)	Any grade n (%)	Grade 1–2 n (%)	Grade ≥ 3 n (%)
Any	264 (85.2)	169 (52.8)	95 (29.7)	103 (64.8)	64 (40.3)	39 (24.5)
Anaemia	92 (28.8)	78 (24.4)	14 (4.4)	33 (20.8)	26 (16.4)	7 (4.4)
Aspartate aminotransferase increased	84 (26.3)	82 (25.6)	2 (0.6)	26 (16.4)	24 (15.1)	2 (1.3)
Alanine aminotransferase increased	80 (25.0)	78 (24.4)	2 (0.6)	29 (18.2)	26 (16.4)	3 (1.9)
Neutrophil count decreased	56 (17.5)	29 (9.1)	27 (8.4)	32 (20.1)	12 (7.5)	20 (12.6)
White blood cell count decreased	53 (16.6)	39 (12.2)	14 (4.4)	26 (16.4)	19 (11.9)	7 (4.4)
Rash	46 (14.4)	43 (13.4)	3 (0.9)	11 (6.9)	11 (6.9)	0
Hypothyroidism	34 (10.6)	34 (10.6)	0	3 (1.9)	3 (1.9)	0
γ-Glutamyltransferase increased	33 (10.3)	27 (8.4)	6 (1.9)	8 (5.0)	6 (3.8)	2 (1.3)
Platelet count decreased	30 (9.4)	20 (6.3)	10 (3.1)	18 (11.3)	13 (8.2)	5 (3.1)
Hyperthyroidism	20 (6.3)	20 (6.3)	0	1 (0.6)	1 (0.6)	0
Leukopenia	12 (3.8)	11 (3.4)	1 (0.3)	4 (2.5)	4 (2.5)	0
Pneumonia	12 (3.8)	10 (3.1)	2 (0.6)	3 (1.9)	1 (0.6)	2 (1.3)
Pyrexia	16 (5.0)	16 (5.0)	0	3 (1.9)	3 (1.9)	0
AE, adverse event; NCI-CTCAE, National Cancer Institute Common Terminology Criteria for Adverse Events.

Objective responses were reported for 203 patients in the sugemalimab group and 64 patients in the placebo group (all partial responses); the ORRs were thus 63.4% and 40.3%, respectively, and were statistically significantly different (P < 0.0001). (Table 1) The change from baseline in the sum of the longest diameters of target lesions is shown in Figure S1. Median duration of response was 9.9 months (95% CI 8.6–13.2) with sugemalimab and 4.4 months (95% CI 3.5–6.1) with placebo (Figure S2).

Efficacy and safety data of patients who received at least 2 years of treatment

At the data cut-off date, 45 (14.1%) patients in the sugemalimab group and six (3.8%) in the placebo group had received sugemalimab or placebo treatment for at least 2 years. Among the patients who were treated with sugemalimab in the double-blind phase for at least 2 years, 33 (73.3%) had PD-L1 ≥ 1% and 29 patients (64.4%) were non-squamous NSCLC; 10 patients (22.2%) had brain metastases and three patients (6.7%) had liver metastases at baseline. The median treatment duration for these 45 patients in the sugemalimab group was 27.3 (IQR 25.2–29.3) months. Forty-one patients had a partial response (91.1%) and four (8.9%) patients had stable disease (Table S8). The majority of these patients (38, 84.4%) were still under study treatment per investigators’ assessment beyond 2 years. Four patients discontinued the treatment due to the maximum treatment of 2 years being reached, two due to radiographic disease progression and one due to withdrawal by patient (Table S8). For patients who were treated with sugemalimab in the double-blind phase for at least 2 years, immune-related AEs were reported for 21 (46.7%) patients with grade ≥ 3 in four (19.0%) patients (Table S8).

Crossover treatment

Of the 45 patients in the placebo group who received the crossover sugemalimab monotherapy, 24 (53.3%) had non-squamous NSCLC and 21 (46.7%) had squamous NSCLC; 27 (60.0%) had PD-L1 expression ≥ 1% and 18 (40.0%) had PD-L1 < 1% (Table S3). After a median follow-up of 17.5 months (IQR 14.3–19.2), median PFS and OS (from the start of the crossover sugemalimab monotherapy) were 4.1 months (95% CI 2.1–5.8) and 14.4 months (95% CI 10.4–18.7), respectively (Figure S3). Five of 45 (11.1%) patients had a confirmed objective response, and median duration of response was 10.4 months (95% CI 3.5-not reached) (Table S9). The safety profile in the crossover phase was consistent with prior studies of sugemalimab monotherapy (Tables S15 and S16).

Safety

The proportions of patients reporting AEs, serious AEs, and grade ≥ 3 treatment-related AEs were generally similar between patients in the sugemalimab and placebo groups (Table 2; Tables S10-S12). Immune-related treatment-emergent AEs were reported for 84 (26.3%) patients in the sugemalimab group [grade ≥ 3 in 17 (5.3%)] and seven (4.4%) in the placebo group (all grade 1–2) (Table S13). The most common (> 1% in either group) were hypothyroidism [n = 35 (10.9%) versus n = 1 (0.6%)], non-severe skin adverse reaction [n = 26 (8.1%) versus n = 2 (1.3%)], hyperthyroidism [n = 24 (7.5%) versus n = 2 (1.3%)], hepatitis [n = 6 (1.9%) versus n = 0], pneumonitis [n = 6 (1.9%) versus n = 2 (1.3%)], diabetes mellitus [n = 5 (1.6%) versus n = 0)], and severe skin reaction [n = 5 (1.6%) versus n = 0]. One (0.3%) death occurred due to immune-related pneumonitis (sugemalimab group).

AEs led to death in 20 (6.3%) patients in the sugemalimab group and nine (5.7%) in the placebo group, most commonly death of unknown aetiology [n = 7 (2.2%) versus n = 3 (1.9%)]. The deaths of 11 (3.4%) patients in the sugemalimab group and two (1.3%) in the placebo group were deemed related to one or more of the study drugs. AEs led to permanent discontinuation of sugemalimab or placebo in 45 (14.1%) and 12 (7.5%) patients, respectively; the most frequently reported in both groups was pneumonia [n = 6 (1.9%) versus n = 3 (1.9%)] (Table S14).

The GEMSTONE-302 study met its key secondary end point, showing significantly prolonged OS with sugemalimab plus platinum-based chemotherapy compared with the placebo combination in patients with previously untreated, metastatic NSCLC with no known driver oncogene alterations. These data represent the first report of a statistically significant and clinically meaningful OS benefit with an anti-PD-L1 monoclonal antibody administered in combination with chemotherapy as a first-line treatment in patients with either metastatic non-squamous or squamous NSCLC (Fig. 2A). The findings of this protocol-specified interim OS analysis support the statistically significant and clinically meaningful improvement in PFS with sugemalimab versus the placebo combination in the GEMSTONE-302 patient population.¹¹ Moreover, the OS benefit was observed across all analysed subgroups, including different pathological tumour types (non-squamous and squamous NSCLC) and different levels of PD-L1 expression (Fig. 2B).

At present, for individuals with metastatic NSCLC lacking a druggable oncogenic driver, the preferred standard of care is pembrolizumab, a PD-1 inhibitor, combined with a platinum-based chemotherapy doublet, with the specific doublet dependent upon whether the individual has squamous or non-squamous histology. This recommendation is supported by the phase III KEYNOTE-189 and KEYNOTE-407 trials. In KEYNOTE-189, which enrolled patients with previously untreated, metastatic, non-squamous NSCLC, median OS was 22.0 months (95% CI 19.5–24.5) for the pembrolizumab plus chemotherapy arm and 10.6 months (95% CI 8.7–13.6) for the placebo plus chemotherapy arm (HR 0.5, 95% CI 0.46–0.69).¹ Notably, OS in the placebo group of KEYNOTE-189 was shorter than the 14–17 months reported in some maintenance trials of pemetrexed.^17,18 In GEMSTONE-302, median OS for the non-squamous population in the placebo group was even longer than those in the pemetrexed maintenance trials at 19.8 months. Despite this, the OS for the sugemalimab group was longer than for the placebo group (median 26.9 months versus 19.8 months, HR 0.72), further demonstrating that the effectiveness of sugemalimab plus chemotherapy is comparable to that of pembrolizumab plus chemotherapy, the globally recognised standard of care for patients with non-squamous NSCLC. In KEYNOTE-407, median OS in patients with previously untreated, metastatic, squamous NSCLC was 17.1 months (95% CI 14.4–19.9) with pembrolizumab and 11.6 months (95% CI 10.1–13.7) without (HR 0.71, 95% CI 0.58–0.88).² That improvement was not well maintained, with 24-month OS rates of 37.5% and 30.6%, respectively.² The OS in KEYNOTE-407 may have been blunted by the high crossover rate and the second-line activity of pembrolizumab in patients with PD-L1-positive, advanced NSCLC, and especially those with PD-L1 ≥ 1%. In the crossover-adjustment analysis using a two-stage method, the HR for OS was 0.59, compared with an HR for OS of 0.71 in the intention-to-treat analysis. In contrast, in GEMSTONE-302, sugemalimab showed a similar, but sustained survival benefit in patients with squamous NSCLC, despite a high subsequent on-study crossover to sugemalimab. Median OS was 23.3 months in the sugemalimab group versus 12.2 months in the placebo group (HR 0.56), with 24-month OS rates of 48.8% and 24.8%, respectively. Adjustment for protocol-specified crossover using the two-stage method showed a further HR improvement for OS in squamous pathology from 0.56 to 0.49. Therefore, sugemalimab combinations provide similar efficacy to pembrolizumab combinations in patients with either non-squamous or squamous metastatic NSCLC.

To date, no PD-L1 inhibitors have shown a statistically significant OS benefit in both first-line non-squamous and squamous NSCLC when used in combination with chemotherapy. In two phase III trials, first-line combination therapy with atezolizumab plus platinum-based chemotherapy in non-squamous (IMpower132) and squamous (IMpower131) NSCLC did not provide a significant improvement in OS compared with the respective placebo combination arms.^{4,5 Moreover, in the recent phase III POSEIDON study, which included patients with either non−squamous or squamous metastatic NSCLC, median OS was not significantly improved by first−line treatment with durvalumab plus platinum−based chemotherapy compared with the placebo combination, irrespective of tumour pathological type.6 GEMSTONE−302 is therefore the first study to show superior OS with a PD−L1 inhibitor, sugemalimab, over placebo, in combination with standard−of−care chemotherapy for both non−squamous and squamous NSCLC (Figure 2B and Figure 3)}.

The present results showed an OS benefit with sugemalimab plus chemotherapy over placebo plus chemotherapy, not only in tumours with PD-L1 ≥ 1%, but also in those with low PD-L1 (< 1%) expression. OS benefits with pembrolizumab combinations were observed both in patients with either PD-L1 ≥ 1% or PD-L1 < 1%.¹⁹ Several trials of chemotherapy combinations with atezolizumab and durvalumab have failed to show improvement in OS in previously untreated NSCLC (non-squamous or squamous) with low (< 1%) tumour PD-L1 expression.^4–6 Other Chinese studies do not show these data in a PD-L1 < 1% population.

A greater reduction in tumour burden was observed in the sugemalimab group over the placebo group in this study, as reflected by the significantly higher ORR (63.4% versus 40.3%; P < 0.0001 for the between-treatment difference of 23.2%). These objective responses (all partial responses) were more durable in the sugemalimab group, with a two-fold increase in duration of response over the placebo group (9.9 months versus 4.4 months). The safety profile of the sugemalimab combination was comparable to that reported in the primary analysis,¹¹ with no new safety signals after the additional 17 + months of follow-up.

There was evidence of meaningful anti-tumour activity among the 45 patients who crossed over to receive sugemalimab monotherapy. Five patients achieved confirmed partial responses, for an ORR of 11.1%, and 20 (44.4%) patients achieved a best overall response of stable disease. Medians of PFS and OS in this crossover group were 4.1 months and 14.4 months, respectively. The efficacy data are generally comparable with the reported data of other PD-1/PD-L1 inhibitors in the second-line setting.²⁰

Sugemalimab in combination with platinum-based chemotherapy had a particular benefit over the placebo combination among patients with brain metastases (Fig. 2B). Until recently, these patients have generally been excluded from clinical trials amid concerns over poor survival and increased risk of toxicity. Despite guidance from the United States Food and Drug Administration recommending that patients with brain metastases now be considered for inclusion in clinical trials,²¹ they were excluded from one-third of registrational trials for recently approved cancer therapies across a variety of indications, including NSCLC.²² The incidence of brain metastases in NSCLC is thought to be increasing.²³ Overall, around 10% of all NSCLC patients and 26% of those with stage IV disease have brain metastases at presentation. They are particularly common among patients harbouring EGFR mutations or ALK alterations and those with a non-squamous pathology.^22,24,25 Patients with brain metastases have poorer outcomes than those who do not.²³ Improvements in the treatment of extracranial disease that prolong survival, and more sophisticated imaging techniques, emphasise the need for improved treatments for brain metastases to further extend survival.^23,25 Inclusion of patients with brain metastases in the present study provides important preliminary data and supports further study in that patient population.

One limitation of this study is the lack of comprehensive analysis of predictive biomarkers and the association between biomarkers and survival benefit with the combination therapy need to be further explored.

In summary, GEMSTONE-302 is the first study to show a statistically significant and clinically meaningful improvement in both PFS and OS with a first-line PD-L1 inhibitor plus chemotherapy combination (versus a placebo plus chemotherapy combination) in patients with metastatic NSCLC with no known driver oncogene alterations, irrespective of their tumour pathological type and PD-L1 expression. Taken together, these findings indicate that combination therapy with sugemalimab plus platinum-based chemotherapy is an effective option with a well-tolerated safety profile for this patient population.

Study design and participants

The design of GEMSTONE-302 study has been described previously.¹¹ Briefly, eligible adult patients were aged 18–75 years with histologically or cytologically confirmed stage IV non-squamous or squamous NSCLC, without known EGFR-sensitising mutations, or ALK, ROS1, or RET fusions, had not received systemic treatment for metastatic disease, and had a baseline Eastern Cooperative Oncology Group performance status (ECOG PS) of 0 or 1, measurable disease per Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST v1.1),¹⁴ and tumour tissue available for PD-L1 expression testing. Patients with brain metastases were not eligible unless the metastases were previously treated or asymptomatic.

Patients were randomised 2:1 to receive sugemalimab (1200 mg) or placebo plus carboplatin (area under the time-concentration curve [AUC] 5 mg/mL/min) and pemetrexed (500 mg/m²) intravenously on day 1 of every 3-week cycle, for up to four cycles, followed by maintenance treatment with pemetrexed (500 mg/m²) plus either sugemalimab (1200 mg) or placebo every 3 weeks (Q3W). Patients with squamous NSCLC received sugemalimab (1200 mg) or placebo plus carboplatin (AUC 5 mg/mL/min) and paclitaxel (175 mg/m²) intravenously on day 1 of every 3-week cycle, for up to four cycles, followed by maintenance therapy with either sugemalimab (1200 mg) or placebo Q3W. Randomisation was stratified by ECOG PS (0 versus 1), histology (non-squamous NSCLC versus squamous NSCLC) and PD-L1 level (≥ 1% versus < 1%). Patients in the placebo group with radiographic evidence of progressive disease were allowed to cross over to receive open-label sugemalimab monotherapy (1200 mg intravenously on day 1 of every 3-week cycle, for up to 35 cycles; henceforth the crossover phase) at the investigator’s discretion, provided they met prespecified criteria (Supplementary methods).

All patients provided written informed consent to participate prior to screening and enrolment. The study conformed with the International Conference on Harmonisation guidelines on Good Clinical Practice, and the Declaration of Helsinki, and was approved by the ethics committees at each participating centre.

Assessment and outcomes

The primary end point was investigator-assessed PFS according to RECIST v1.1¹⁴ in the intention-to-treat population. Secondary end points (per RECIST V1.1) included OS, investigator-assessed PFS in patients with PD-L1 ≥ 1%, PFS assessed by blinded independent central review (BICR), investigator-assessed ORR and duration of response (DoR). Confirmed investigator-assessed ORR, duration of response, PFS according to RECIST v1.1, and OS were also evaluated for all patients in the crossover phase. Safety of sugemalimab was also assessed.

Tumour samples (collected at or after diagnosis) were assessed for PD-L1 expression at a central laboratory. Tumour assessments for treatment response were conducted at baseline, at Weeks 6 and 12, and then every 9 weeks for the first year, and every 12 weeks thereafter until disease progression, loss to follow-up, death or study completion (whichever occurred first). Patients were followed up for survival every 12 weeks after the last treatment dose. Treatment-emergent adverse events and serious adverse events were graded according to the US National Cancer Institute Common Terminology Criteria for Adverse Events version 4.03.¹⁵

Statistical analysis

A full account of the statistical methodology has been published¹¹ and is presented in the Supplementary methods, along with sample size calculations and definitions of analysis populations (Supplementary methods, Tables S1 and S2). The cut-off date for all analyses reported here (OS, PFS in ITT population and in patients with PD-L1 ≥ 1%, and ORR) was 22 November 2021. Kaplan–Meier methodology was used to estimate time-to-event parameters, and stratified and unstratified log-rank tests were used to calculate P values. Stratified and unstratified Cox model were used to estimate HRs with 95% CIs. ORRs were compared between groups using stratified Mantel–Haenszel tests. A sequential testing method was used to control overall type I error. The Lan–DeMets method with an approximate Pocock boundary was used to calculate the P value boundary (two-sided 0.0396) for interim analysis of OS.¹⁶ Analyses of OS adjusting for protocol-specified treatment crossover or receiving subsequent non-study PD-1/PD-L1 anti-cancer therapy for patients in the placebo group were conducted using the inverse probability of censoring weighting (IPCW), rank-preserving structural-failure time model (RPSFTM), and two-stage accelerated-failure time model (AFTM). In addition, OS analyses adjusting for the protocol-specified treatment crossover only were conducted based on two-stage AFTM for the total patient population and by pathological types.

Acknowledgements

The authors are grateful to all of the patients and their families for participating in the study, as well as the study investigators and other staff. We also thank Shuo Chi from CStone Pharmaceuticals and Xin Wang from IQVIA (funded by CStone Pharmaceuticals) for assistance with the statistical analysis. Medical writing assistance was provided by Jacqueline Kolston and Susan Neville of Parexel (funded by CStone Pharmaceuticals), and Mingjue Zhao and Aijia Wu of CStone Pharmaceuticals.

Funding

This work was supported by CStone Pharmaceuticals.

Declaration of interests

CZ reports speaker honoraria from CStone Pharmaceuticals during the course of this study and, outside of this study, reports speaker honoraria from Lilly China, Sanofi, Boehringer Ingelheim, Roche, MSD, Qilu Pharmaceutical, Hengrui Therapeutics, Innovent Biologics, Luye Pharma Group, TopAlliance Biosciences, and Amoy Diagnostics; and is an advisor for Innovent Biologics, Hengrui Therapeutics, Qilu Pharmaceutical, and TopAlliance Biosciences. JW, RC, MQ, HW, and JY are paid employees of CStone Pharmaceuticals and JW and JY declare stock ownership in the company. All remaining authors have declared no conflicts of interests.

Contributors

CZ, ZW, YL, ZM, HS, JW, and JY provided substantial contributions to the conception and design of the study. CZ, ZW, MS, LC, ZM, RW, YY, WY, SS, JCh, WZ, JCu, XC, YLiu, HS, CH, JLiu, MW, XL, PS, YS, JZ, JLi, KG, CW, HZ, YZ, and CL enrolled and treated patients. HW did the statistical analyses and generated data. CZ, JW, RC, MQ, and JY analysed and interpreted the data. CZ, JW, RC, MQ, and JY drafted and provided critical revision of the manuscript. CZ, JW, RC, MQ, and HW verified the data. All authors had full access to all the data in the study, reviewed the data, contributed to the development of the manuscript, approved the final version, and had final responsibility for the decision to submit for publication.

Rodríguez-Abreu D, Powell SF, Hochmair MJ, et al. Pemetrexed plus platinum with or without pembrolizumab in patients with previously untreated metastatic nonsquamous NSCLC: protocol-specified final analysis from KEYNOTE-189. Ann Oncol 2021;32:881–895.
Paz-Ares L, Vicente D, Tafreshi A, et al. A randomized, placebo-controlled trial of pembrolizumab plus chemotherapy in patients with metastatic squamous NSCLC: protocol-specified final analysis of KEYNOTE-407. J Thorac Oncol 2020;15:1657–1669.
Gray J, Rodríguez-Abreu D, Powell SF, et al. FP13.02 pembrolizumab + pemetrexed-platinum vs pemetrexed-platinum for metastatic NSCLC: 4-year follow-up from KEYNOTE-189. J Thorac Oncol 2021;16:S224.
Nishio M, Barlesi F, West H, et al. Atezolizumab plus chemotherapy for first-line treatment of nonsquamous NSCLC: results from the randomized phase 3 IMpower132 trial. J Thorac Oncol 2021;16:653–664.
Jotte R, Cappuzzo F, Vynnychenko I, et al. Atezolizumab in combination with carboplatin and nab-paclitaxel in advanced squamous NSCLC (IMpower131): results from a randomized phase III trial. J Thorac Oncol 2020;15:1351–1360.
Johnson M, Cho BC, Luft A, et al. PL02.01 durvalumab ± tremelimumab + chemotherapy as first-line treatment for mNSCLC: results from the phase 3 POSEIDON study. J Thorac Oncol 2021;16:S844.
Baggstrom MQ, Socinski MA, Wang XF, et al. Maintenance sunitinib following initial platinum-based combination chemotherapy in advanced-stage IIIB/IV non-small cell lung cancer: a randomized, double-blind, placebo-controlled phase III stud –CALGB 30607 (Alliance). Journal of Thoracic Oncology 2017;12:843–849.
Wang J, Lu S, Yu X, et al. Tislelizumab plus chemotherapy vs chemotherapy alone as first-line treatment for advanced squamous non–small-cell lung cancer: a phase 3 randomized clinical trial. JAMA Oncology 2021;7:709–717.
Ren S, Chen J, Xu X, et al. Camrelizumab plus carboplatin and paclitaxel as first-line treatment for advanced squamous NSCLC (CameL-Sq): a phase 3 trial. Journal of Thoracic Oncology 2022;17:544–557.
Gong J, Cao J, Zhang Q, et al. Safety, antitumor activity and biomarkers of sugemalimab in Chinese patients with advanced solid tumors or lymphomas: results from the first-in-human phase 1 trial. Cancer Immunol Immunother 2022;Preprint: 10.1007/s00262-021-03102-3.
Zhou C, Wang Z, Sun Y, et al. Sugemalimab versus placebo, in combination with platinum-based chemotherapy, as first-line treatment of metastatic non-small-cell lung cancer (GEMSTONE-302): interim and final analyses of a double-blind, randomised, phase 3 clinical trial. Lancet Oncol 2022;23:220–233.
CStone Pharmaceuticals (Suzhou). CStone announces the GEMSTONE-302 study of sugemalimab met the endpoint of overall survival in the first-line treatment of metastatic non-small cell lung cancer patients. Available at: https://www.cstonepharma.com/en/html/news/2738.html. Accessed Mar 24, 2022.
CStone Pharmaceuticals (Suzhou) Co., Ltd. Cejemly® (sugemalimab) injection, for intravenous use (in Chinese). Available at: https://www.cstonepharma.com/uploads/2022/01/164197952116940.pdf. Accessed Feb 3, 2022.
Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228–247.
U.S. Department of Health and Human Services, National Institutes of Health, National Cancer Institute. Common terminology criteria for adverse events (CTCAE), Version 4.03. Available at: https://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03/CTCAE_4.03_2010-06-14_QuickReference_8.5x11.pdf. Accessed Apr 05, 2022.
Chen LM, Ibrahim JG, Chu H. Flexible stopping boundaries when changing primary endpoints after unblinded interim analyses. J Biopharm Stat 2014;24:817–833.
Paz-Ares LG, de Marinis F, Dediu M, et al. PARAMOUNT: final overall survival results of the phase III study of maintenance pemetrexed versus placebo immediately after induction treatment with pemetrexed plus cisplatin for advanced nonsquamous non-small-cell lung cancer. J Clin Oncol 2013;31:2895–2902.
Scagliotti G, Brodowicz T, Shepherd FA, et al. Treatment-by-histology interaction analyses in three phase III trials show superiority of pemetrexed in nonsquamous non-small cell lung cancer. J Thorac Oncol 2011;6:64–70.
Gadgeel S, Rodríguez-Abreu D, Speranza G, et al. Updated analysis from KEYNOTE-189: pembrolizumab or placebo plus pemetrexed and platinum for previously untreated metastatic nonsquamous non–small-cell lung cancer. Journal of Clinical Oncology 2020;38:1505–1517.
Qu J, Mei Q, Liu L, et al. The progress and challenge of anti-PD-1/PD-L1 immunotherapy in treating non-small cell lung cancer. Ther Adv Med Oncol 2021;13:1758835921992968.
U.S. Department of Health and Human Services. Cancer clinical trial eligibility criteria: brain metastases guidance for industry. Available at: https://www.fda.gov/media/121317/download. Accessed Apr 5, 2022.
Tan AC, Boggs DH, Lee EQ, et al. Clinical trial eligibility criteria and recently approved cancer therapies for patients with brain metastases. Front Oncol 2022;11:780379.
Lim JH, Um SW. The risk factors for brain metastases in patients with non-small cell lung cancer. Ann Transl Med 2018;6:S66.
Rangachari D, Yamaguchi N, VanderLaan PA, et al. Brain metastases in patients with EGFR-mutated or ALK-rearranged non-small-cell lung cancers. Lung Cancer 2015;88:108–111.
Waqar SN, Samson PP, Robinson CG, et al. Non-small-cell lung cancer with brain metastasis at presentation. Clin Lung Cancer 2018;19:e373-e379.

Supplementary Materials are not available with this version.

Yes there is potential Competing Interest. CZ reports speaker honoraria from CStone Pharmaceuticals during the course of this study and, outside of this study, reports speaker honoraria from Lilly China, Sanofi, Boehringer Ingelheim, Roche, MSD, Qilu Pharmaceutical, Hengrui Therapeutics, Innovent Biologics, Luye Pharma Group, TopAlliance Biosciences, and Amoy Diagnostics; and is an advisor for Innovent Biologics, Hengrui Therapeutics, Qilu Pharmaceutical, and TopAlliance Biosciences. JW, RC, MQ, HW, and JY are paid employees of CStone Pharmaceuticals and JW and JY declare stock ownership in the company. All remaining authors have declared no conflicts of interests.

Download PDF

Journal Publication

published 15 Jun, 2023

Read the published version in Nature Cancer →

Version 1

posted

You are reading this latest preprint version

A protocol pre-specified interim overall survival analysis of the randomised phase III GEMSTONE-302 trial: sugemalimab or placebo plus platinum-based chemotherapy for first-line metastatic non-small-cell lung cancer

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Results

Patients and treatments

Overall survival

Progression free survival

Efficacy and safety data of patients who received at least 2 years of treatment

Crossover treatment

Safety

Discussion

Conclusions

Methods

Study design and participants

Assessment and outcomes

Statistical analysis

Declarations

References

Supplementary Materials

Additional Declarations

Status:

Journal Publication

Version 1