Nimotuzumab Plus Gemcitabine for K-Ras Wild-Type Locally Advanced or Metastatic Pancreatic Cancer

K-Ras pancreatic cancers benefit overall survival in phase 3 NOTABLE study.

Abstract

PURPOSE

In a phase IIb trial of nimotuzumab plus gemcitabine, substantial clinical benefits were observed in patients with locally advanced or metastatic pancreatic cancer (PC). Therefore, we conducted a phase III clinical study to verify the efficacy and safety of this combination regimen in patients with K-Ras wild-type tumors (ClinicalTrials.gov identifier: NCT02395016).

PATIENTS AND METHODS

Eligible patients were randomly assigned to receive nimotuzumab (400 mg once per week) or placebo followed by gemcitabine (1,000 mg/m² on days 1, 8, and 15, once every 4 weeks) until disease progression or unacceptable toxicity. The primary end point was overall survival (OS) and the secondary end points were progression-free survival (PFS), response rates, and safety.

RESULTS

A total of 480 patients were screened; 92 patients were enrolled and 82 patients with K-Ras wild-type tumors were eligible. In the full analysis set, the median OS was 10.9 versus 8.5 months, while the restricted mean survival time (RMST) was 18.05 versus 11.14 months for the investigational versus control arm (ratio of control v investigation = 0.62 [0.40-0.97]; P = .036). Median PFS was 4.2 versus 3.6 months in the investigational versus control arm (log-rank P = .04; hazard ratio, 0.60 [0.37-0.99]) and the restricted mean PFS time was 8.08 versus 4.76 months (RMST ratio, 0.58 [0.38-0.90]; P = .036). Both OS and PFS were longer in the nimotuzumab group than in the placebo group. The objective response rates and disease control rates were 7% versus 10% and 68% versus 63% for the investigational and control groups, respectively. The incidence of adverse events were comparable between the two groups.

CONCLUSION

In patients with locally advanced or metastatic K-Ras wild-type PC, nimotuzumab plus gemcitabine significantly improved OS and PFS with a good safety profile.

INTRODUCTION

Pancreatic cancer (PC) is a common and fatal malignancy that occurs worldwide. Eighty percent of PC cases are diagnosed as advanced or metastatic stage.¹ Twenty-five percent of patients with advanced PC are at a locally advanced stage, with a median overall survival (mOS) of 6-9 months. Approximately 60%-70% of patients with advanced PC are at the metastatic stage, with mOS of only 3-5 months.^2–9 Unfortunately, the current regimens are unsatisfactory. Encouraging the achievement of new regimens has been developed in many other cancer fields but rarely in PC. Therefore, there is an urgent need for effective therapies to prolong the survival of patients with PC. Nimotuzumab is a recombinant humanized monoclonal antibody against the epidermal growth factor receptor (EGFR). A phase IIb study (PCS07) done in Germany showed that nimotuzumab plus gemcitabine might significantly improve the mOS (8.6 months v 6.0 months) of locally advanced or metastatic PC, especially for the wild-type K-Ras gene subgroup (11.62 months v 5.67 months; 1-year survival rates, 53.8% v 15.8%, respectively).¹⁰ In 2014, gemcitabine was widely used and was considered the standard of care for first-line treatment regimen for patients with advanced PC in China. Thus, we conducted a phase III trial (NOTABLE study) to verify the efficacy and safety of nimotuzumab plus gemcitabine as first-line treatment for patients with locally advanced or metastatic PC.

CONTEXT

• Key Objective

• Does nimotuzumab (a humanized epithelial growth factor receptor monoclonal antibody) plus gemcitabine regimen provide an overall survival (OS) benefit for patients with locally advanced or metastatic K-Ras wild-type pancreatic cancer (PC)?

• Knowledge Generated

• In a multicenter, randomized, double-blinded, phase III clinical trial in China, patients with locally advanced or metastatic K-Ras wild-type PC were enrolled. The results showed that nimotuzumab plus gemcitabine significantly improved OS and progression-free survival with a good safety profile compared with the placebo-gemcitabine group.

• Relevance (E.M. O'Reilly)

• This randomized trial confirms that identification of KRAS wild-type PC is clinically important and therapeutically relevant and can be successfully targeted.*

  *Relevance section written by JCO Associate Editor Eileen M. O'Reilly, MD.

PATIENTS AND METHODS

Study Oversight

The NOTABLE study was approved by each institution's independent ethics committee, the Chinese Drug Regulatory Authority, and the National Medical Products Administration, adhering to the Helsinki Declaration on human medical research.¹¹

The investigators and authors designed the Protocol (online only) using a sponsor (Biotech Pharmaceutical Co, Ltd). The principal investigators from each site adhered to the study protocol. In addition, data were collected and analyzed by professional third-party statisticians who certified the accuracy and completeness of the reported data.

Patents

Patients with locally advanced or metastatic PC were screened. Each patient signed an informed consent form before screening. Eligible patients were age 18-75 years with histologically confirmed locally advanced/metastatic PC with the K-Ras wild-type gene (Sanger gene sequence). Other inclusion criteria were a Karnofsky performance status (KPS) score ≥60, ineligibility for radical radiotherapy or surgery, previous adjuvant chemotherapy (if any) for more than 6 months before enrollment, at least one measurable lesion (RECIST 1.1),¹² and an estimated life expectancy of 12 weeks or more, together with normal organ function.

Study Design and Treatment

This was a prospective, multicenter, randomized, double-blinded, phase III clinical study (NOTABLE study), in which eligible patients were randomly assigned (1:1) to receive intravenous nimotuzumab (400 mg) or a placebo once a week. Gemcitabine (1,000 mg/m²) was administered intravenously once per week (on days 1, 8, and 15) for 3 weeks, rest for 1 week, then once per week for 4 weeks for one treatment cycle. Treatment was continued until disease progression, unmanageable toxicity, or withdrawal of consent was observed. The patients were followed up until death or study completion.

Patients were stratified by tumor localization (head, body, or tail of the pancreas) according to previous surgery, history of biliary obstruction treatment, and adjuvant chemotherapy.

Assessments

The investigators evaluated the tumor response via computed tomography or magnetic resonance imaging scan every 8 weeks, according to RECIST version 1.1. The clinical benefit response (CBR) was evaluated every 8 weeks on the basis of the Burris criteria.¹³ Adverse events (AEs) and serious adverse events were monitored by the investigators from the beginning of the treatment and up to 30 days after treatment completion. The detected events were graded according to the Common Terminology Criteria for Adverse Events, version 4.03,¹⁴ and coded and summarized according to the preferred terms in the Medical Dictionary for Regulatory Activities, version 24.0.

Study End Points

The primary end point was overall survival (OS), defined as the time from random assignment to death due to any cause. Subgroup analyses of OS were based on prespecified stratification factors. The secondary end points were progression-free survival (PFS), defined as the time from random assignment until objective tumor progression or death; time to progression (TTP, defined as the time from random assignment until objective tumor progression); objective response rate (ORR), including complete response (CR) and partial response (PR); disease control rate (DCR), including CR, PR, and stable disease; and CBR on the basis of changes in pain, KPS, and weight. Treatment-related adverse events (TRAEs) were defined as adverse drug reactions (ADRs).

Statistical Analysis

The primary end point was evaluated in the full analysis set (FAS; all eligible patients who received at least one dose of nimotuzumab and had one evaluation of efficacy). Secondary end points were analyzed only in the FAS. A prespecified subgroup analysis of OS and PFS was performed using the FAS. Safety analysis was performed using the safety analysis set (SS; patients received at least one antibody dose).

Statistical analyses were performed using the SAS version 9.4 (SAS Institute, Cary, NC). OS was the primary efficacy end point and was analyzed using the Kaplan-Meier method and stratified log-rank test. If the proportional hazards assumption was unsatisfactory (eg, survival curve crossover), the restricted mean survival time (RMST) model¹⁵ was used to analyze the survival benefit difference. RMST is suggested as a novel alternative measure in survival analyses and may be useful when the proportional hazards assumption cannot be made. A right-censored pattern was used for the OS data. The time point of the RMST model was the closest to the end of the trial.

We calculated that, with a sample of 276 patients, the study would have 80% power to detect a hazard ratio (HR) for death with nimotuzumab plus gemcitabine (investigational arm, or nimotuzumab-gemcitabine) versus placebo plus gemcitabine (control arm, or placebo-gemcitabine) of 0.69. This estimation was based on the results of the full population in the PCS07 study, where the median OS of the gemcitabine monotherapy group was 6 months and the median OS of the investigational arm was 8.7 months. Since patients with K-Ras wild-type PC were rare and recruitment was difficult, we calculated that with a sample of 79 patients and 64 events, the study would have 80% power to detect a HR for death with nimotuzumab-gemcitabine versus placebo-gemcitabine. This assumption was made on the basis of the result of patients with K-Ras wild-type PC in PCS07 study, where the median OS of gemcitabine monotherapy group was 5.67 months, and the median survival of the investigational group was 11.62 months. The sample size re-estimation was approved by the Chinese regulatory authorities. At the time of approval, 92 patients were enrolled and the study was terminated.

Multivariate analysis of survival data was conducted using the Cox proportional hazard model according to prespecified stratification factors to evaluate the treatment effect. In addition to the HR as a measurement of the relative risk of an event, the RMST ratio between arms (control v investigation) was estimated after adjusting for covariates. For this analysis, a ratio of <1 indicated survival improvement in the investigational arm. Unlike the HR, the RMST ratio does not rely on any model assumption.¹⁶

The ORR was analyzed using the Fisher test, while the DCR and CBR were evaluated using the chi-squared test. AEs were classified according to the system organ class, preferred term, or severity grade.

RESULTS

Patients and Treatment Groups

A total of 480 patients from 25 study sites in China were screened between April 2015 and March 2021. Then, 92 patients with the K-Ras wild-type gene were eligible for random assignment, and 90 patients were allocated for treatment. Eight patients who did not have K-Ras wild-type tumors according to a gene confirmatory procedure (seven cases) or PC (one gallbladder carcinoma case), detected during monitoring, were excluded. Finally, 82 patients were analyzed as part of the FAS, and 41 were included in each group. The FAS group was the modified intent-to-treat group (Fig 1).

FIG 1.

CONSORT diagram. FAS, full analysis set; gem, gemcitabine; SS, safety set.

The demographic and clinical characteristics at baseline were well balanced between the two groups (Table 1).

TABLE 1.

Characteristics of the Patients at Baseline

Efficacy

OS

The OS analysis was based on 76 deaths, including 36 deaths in the nimotuzumab-gemcitabine group and 40 deaths in the placebo-gemcitabine group. The mean follow-up time was 57.6 months and 16.6 months, respectively, for the two groups. The data cutoff was January 27, 2022. The median OS was 10.9 months (95% CI, 5.6 to 16.3) in the nimotuzumab-gemcitabine group and 8.5 months (95% CI, 5.7 to 10.0) in the placebo-gemcitabine group (Cox proportional HR, 0.66; 95% CI, 0.42 to 1.05; log-rank P = .08; Fig 2A). When using RMST model, the RMST was 18.05 months (95% CI, 11.71 to 24.38) in the nimotuzumab-gemcitabine group and 11.14 months (95% CI, 8.07 to 14.20) in the placebo-gemcitabine group (RMST ratio of control v treatment, 0.62; 95% CI, 0.40 to 0.97; P = .036; Table 2). OS was significantly longer in the nimotuzumab-gemcitabine group (Fig 2C).

FIG 2.

Kaplan-Meier curves for OS and PFS, the RMST test for OS time until the end of trial (75 months). (A) OS. (B) PFS. (C) OS time by the RMST test (arm = 1 is for the nimotuzumab-gemcitabine group, arm = 0 is for the placebo-gemcitabine group; the estimate of RMST is the red area, and the restricted mean lost time is the orange area). (D) PFS time by the RMST test (arm = 1 is for the nimotuzumab-gemcitabine group, arm = 0 is for the placebo-gemcitabine group; estimate of RMST is the red area and the restricted mean lost time is the orange area). OS, overall survival; PFS, progression-free survival; RMST, restricted mean survival time.

TABLE 2.

OS, PFS, and Response Rates

PFS and TTP

PFS was significantly longer in the nimotuzumab-gemcitabine group than in the placebo-gemcitabine group, with median survival time of 4.2 months (95% CI, 2.7 to 7.3) versus 3.6 months (95% CI, 2.0 to 5.0; HR, 0.60; 95% CI, 0.37 to 0.99; log-rank P = .04), respectively (Fig 2B). At a follow-up of 31 months, the restricted mean PFS time was 8.08 months (95% CI, 5.20 to 10.94) in the nimotuzumab-gemcitabine group and 4.76 months (95% CI, 3.41 to 6.09) in the placebo-gemcitabine group (RMST ratio, 0.58; 95% CI, 0.38 to 0.90; P = .036), with a difference of 3.3 months (95% CI, 0.30 to 6.24; adjusted stratification of P = .031; Fig 2D).

The median TTP was 4.7 months (95% CI, 2.7 to 9.0) in the nimotuzumab-gemcitabine group and 3.7 months (95% CI, 2.0 to 5.4) in the placebo-gemcitabine group (HR, 0.67; 95% CI, 0.39 to 1.15; log-rank P = .311) of no significant difference.

ORR and DCR

ORRs were 7.3% (95% CI, 1.5 to 19.9) versus 9.8% (95% CI, 2.7 to 23.1) in the investigational versus control group, and the DCRs were 68.3% (95% CI, 51.9 to 81.9) versus 63.4% (95% CI, 46.9 to 77.9), respectively. The ORR and DCR results were not significantly different between the two groups.

CBR

The CBRs in the nimotuzumab-gemcitabine and placebo-gemcitabine groups were 39.3% (95% CI, 21.2 to 57.4) and 32.2% (95% CI, 15.8 to 48.8), respectively. However, this difference was not statistically significant.

Subgroup Analyses

In the prespecified subgroups, the results were mostly consistent with the pooled results. Compared with the placebo-gemcitabine group, the nimotuzumab-gemcitabine group showed trends of larger survival as well as progression risk reduction in several subgroups, especially for patients without previous biliary obstruction (OS, 11.9 months v 8.4 months; HR, 0.54; 95% CI, 0.33 to 0.89; log-rank P = .035; PFS, 5.5 months v 3.4 months; HR, 0.5; 95% CI, 0.29 to 0.85; log-rank P = .035; Figs 3A and 3B).

FIG 3.

Forest plots of the treatment effect on (A) OS and (B) PFS in prespecified subgroups. The KPS score ranges from 0 to 100, with a higher score indicating better performance status. The HR and P value were obtained under the Cox regression model. HR, hazard ratio; KPS, Karnofsky performance status; OS, overall survival; PFS, progression-free survival.

Treatment Exposure

The median duration of treatment was 16 weeks (range, 1-127 weeks) in the nimotuzumab-gemcitabine group and 13 weeks (range, 1-80 weeks) in the placebo-gemcitabine group, in the FAS population. The median exposure dose of nimotuzumab was 6,400 mg in the nimotuzumab-gemcitabine group (range, 400-50,800 mg; dose frequency, once per week).

Safety

The incidences of AEs in this trial are shown in Table 3. The occurrence of ADRs was similar between the two groups. There was only one case of serious ADR in the nimotuzumab-gemcitabine group. The patient was hospitalized for a CTCAE grade 3 bilirubin increase and anemia, which are listed in the nimotuzumab package insert. Neutropenia, thrombocytopenia, leukopenia, and AST or ALT increased, and rash, fatigue, and fever were comparable between the two groups. No grade 4-5 AEs were observed in this trial.

TABLE 3.

The Overall Incidence of AEs in This Trial

DISCUSSION

This prospective, randomized, controlled, double-blind, multicenter clinical trial demonstrated that nimotuzumab combined with gemcitabine prolonged the median OS time by 2.4 months and decreased the mortality risk by 34% compared with the placebo group by Cox regression. Once the trial was completed, we observed that the survival curves overlapped until 6 months and gradually separated thereafter. In this situation, the proportional hazard assumption for the Cox regression model does not hold. Consequently, it is not a robust method. Subsequently, we used the RMST model to calculate the survival differences between the two curves using the RMST ratio, which behaves as an analog of the Cox proportional HR. The mortality risk decreased by 38%, which was equivalent to RMST ratio 0.62 (0.40-0.97) calculated at the placebo-gemcitabine versus nimotuzumab-gemcitabine direction, and the differences were statistically significant. The RMST for both OS and PFS was significantly longer in the nimotuzumab-gemcitabine group than in the control group. The 1-year and 3-year OS rates were 43.6% versus 26.8% and 13.9% versus 2.7% in the investigational and control arms, respectively. These results demonstrated that nimotuzumab increased the median survival time of patients with advanced PC with K-Ras wild-type tumors.

Compared with other studies, the NOTABLE trial showed that nimotuzumab might confer a remarkable survival benefit. In a previous study, the median OS of gemcitabine monotherapy was 5.65 months.¹³ However, the median OS was increased to 6.24 months when combined with erlotinib,¹⁷ and 8.5 months and 1-year OS rate of 35% when albumin-paclitaxel was added to gemcitabine.¹⁸ In addition to these regimens, the fluorouracil, leucovorin, irinotecan, and oxaliplatin (FOLFIRINOX) schedule had the longest median survival time of 11.1 months; its usage is limited owing to side effects.¹⁹ Compared with the current treatment alternatives, the median OS of nimotuzumab-gemcitabine in this study was 10.9 months, and the 1-year OS rate was 43.6%, which was equal to that of the FOLFIRINOX regimen, but with a better safety profile.

PFS was also significantly higher in the nimotuzumab arm, according to the log-rank P value and RMST method. The ORRs of the two treatment groups were similar; however, the DCR and CBR were higher in the nimotuzumab group, although the difference was not statistically significant. These results indicate that nimotuzumab may improve disease control and slow the progression of advanced PC.

In the subgroup analyses, the nimotuzumab-gemcitabine combination showed a greater benefit, which was consistent with the results of the total population analysis. The subgroup analysis should be interpreted with caution because after adjusting the sample size, the subgroups were small. Regarding the safety profile, AEs or ADRs were comparable between the two groups and did not increase after the addition of nimotuzumab. The frequency of total hematologic or nonhematologic AEs (including grade 3 AEs) >10% was similar between the two groups. All TRAEs were grade 1-3, and no grade 4-5 adverse reactions or events occurred. Only grade 1-2 rash reactions were observed. This trial confirmed that nimotuzumab has a good safety profile.

The proportion of K-Ras wild-type tumors in our study was 17.1%. This percentage was higher than the internationally reported data, but similar to a previous nationwide study in China (16.8%).²⁰

One limitation of this study is that EGFR expression and other genetic mutation types were not evaluated. It is unknown whether patients with activation of the mitogen-activated protein kinase pathway because of BRAF mutations would benefit from nimotuzumab treatment. Alternatively, other kinase fusion genes might confer sensitivity to nimotuzumab. Delineating the molecular subtypes of K-Ras wild-type tumors could improve patient selection for nimotuzumab treatment.

Information on subsequent treatments after disease progression was not collected, which is another limitation of this study. Third, although there were no statistically significant differences in baseline characteristics between the two arms, nimotuzumab-treated patients were slightly younger and had a better performance status. These features may have led to the improved treatment compliance. Finally, the study protocol was formulated in 2014, when the standard of care of first-line treatment regimens in China was gemcitabine monotherapy. At that time, FOLFIRINOX was rarely used in China because of its high toxicity (some populations cannot tolerate combination chemotherapy), and gemcitabine/nab-paclitaxel was not the standard practice.

In conclusion, the combination of nimotuzumab and gemcitabine was safe and increased the OS and PFS of patients with locally advanced or metastatic PC with wild-type KRAS.

DATA SHARING STATEMENT

All the authors agree to provide a data-sharing statement for this manuscript. Data-sharing requests should be sent to qinsk@csco.org.cn for deidentified data. Such requests will be considered by the study team after publication, after the review and approval of proposals and with appropriate data-sharing agreements in place.

AUTHOR CONTRIBUTIONS

Conception and design: Shukui Qin, Yuxian Bai, Zishu Wang, Zhendong Chen, Ruihua Xu, Jianming Xu, Jia Chen, Lin Yang, Haijun Zhong, Lin Shen, Chunyi Hao, Deliang Fu, Jianwei Yang, Qiong Wang, Baoli Qin, Hongming Pan, Xianhong Bai, Qingshan Zheng

Administrative support: Lin Shen

Provision of study materials or patients: Jin Li, Ruihua Xu, Lin Yang, Lin Shen, Deliang Fu, Ying Cheng

Collection and assembly of data: Shukui Qin, Jin Li, Yuxian Bai, Zishu Wang, Zhendong Chen, Jia Chen, Ying Yuan, Tianshu Liu, Lin Yang, Haijun Zhong, Donghui Chen, Lin Shen, Chunyi Hao, Deliang Fu, Ying Cheng, Jianwei Yang, Qiong Wang, Baoli Qin, Hongming Pan, Jun Zhang, Qingshan Zheng

Data analysis and interpretation: Shukui Qin, Yuxian Bai, Zishu Wang, Zhendong Chen, Ruihua Xu, Jianming Xu, Hongmei Zhang, Jia Chen, Lin Yang, Haijun Zhong, Donghui Chen, Chunyi Hao, Jianwei Yang, Qiong Wang, Baoli Qin, Hongming Pan, Xianhong Bai, Qingshan Zheng

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

Nimotuzumab Plus Gemcitabine for K-Ras Wild-Type Locally Advanced or Metastatic Pancreatic Cancer

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).