Introduction

Indolent non-Hodgkin lymphoma (iNHL) is mostly of B-cell origin [1] and is generally highly responsive to initial therapy, but most patients experience relapse and require multiple lines of treatment [1,2,3,4]. The most common types of iNHL are follicular lymphoma (FL) and marginal zone lymphoma (MZL), and the proportion of FL (15%/14%) and MZL (8%/8%) in NHL is similar in US/Japanese populations [5]. Incidence rates of NHL, including FL and MZL, in Japan have more than doubled since the early 1990s [5], and although survival for patients with iNHL has improved, relapse continues to be a problem [6].

Currently, there is a broad range of treatment options for patients with relapsed/refractory (R/R) FL, with different efficacy/toxicity profiles [7,8,9]. Patients with R/R MZL are managed using the same options as in patients with R/R FL, with the exception of those with the localized mucosa-associated lymphoid tissue (MALT) subtype [10]. Rituximab is approved as monotherapy by the US Food and Drug Administration and the Japan’s Pharmaceuticals and Medical Devices Agency for CD20 + B-cell NHL [11, 12] and re-treatment with rituximab monotherapy is an option for patients after relapse on a rituximab-containing therapy [8, 12]. Lenalidomide is an IMiD immunomodulatory agent that causes degradation of the transcription factors Aiolos and Ikaros [13, 14] leading to anti-lymphoma and immune-modulation activities. In preclinical studies, lenalidomide restored the response of tumor-infiltrating lymphocytes [15] and increased natural killer cell count and functional activity [16, 17]. Combining lenalidomide + rituximab (R2) has resulted in enhanced antibody-dependent cell-mediated cytotoxicity, immune synapse formation, monocyte-mediated killing, and direct cytotoxicity against FL cells compared with either treatment alone [18].

R2 has shown high response rates in multiple phase II studies in patients with R/R and previously untreated iNHL [16, 19,20,21,22,23]. R2 has also shown similar efficacy and favorable safety profile compared to rituximab-containing chemotherapy in a phase III study of patients with advanced, previously untreated FL [24].

The goal of the AUGMENT phase III study was to prospectively compare the efficacy and safety of R2 vs rituximab + placebo (R-placebo) in patients with R/R iNHL. Recently, results of the AUGMENT study showed statistically superior efficacy, as measured by the primary endpoint—progression-free survival (PFS)—and other secondary endpoints, of R2 vs R-placebo in patients with R/R iNHL [25]. Considering the different treatment landscapes for iNHL globally, subgroup analysis was performed in Japanese patients enrolled in the AUGMENT study, and the results are presented here.

Patients and methods

Eligibility criteria

Patients with MZL or FL (grades 1–3a [transformed FL excluded]) requiring treatment per investigator assessment and aged ≥ 18 years were eligible. Eligibility also included previous treatment with ≥ 1 prior chemotherapy, immunotherapy, or chemoimmunotherapy and ≥ 2 previous doses of rituximab; documented relapsed, refractory, or progressive disease (PD), and not rituximab-refractory. Full details of all inclusion and exclusion criteria are reported in the primary AUGMENT publication [25].

Treatment protocol

A brief summary of the patient population and study method is provided here; for full details, please refer to the primary AUGMENT publication [25]. This was a single-country analysis from the AUGMENT trial, a multicenter, double-blind, randomized, phase III trial designed to evaluate the efficacy and safety of R2 vs R-placebo in Japanese patients with R/R iNHL (NCT01938001). Patients were randomly assigned in a 1:1 ratio to receive R2 or R-placebo and stratified according to previous rituximab treatment (yes or no), time since last anti-lymphoma therapy (≤ 2 vs > 2 years), and histology (FL vs MZL). Treatment was given for a maximum of 12 cycles or until relapse, PD, withdrawal of consent, or unacceptable toxicity. R2 dosing consisted of oral lenalidomide 20 mg/day (10 mg if creatinine clearance was 30–59 mL/min) on days 1–21, plus intravenous rituximab 375 mg/m2 on days 1, 8, 15, and 22 of cycle 1 and day 1 of cycles 2–5, in 28-day cycles. R-placebo was given on the same dosing schedule. Upon treatment completion or discontinuation, patients were followed for efficacy, subsequent anti-lymphoma therapies, and second primary cancers for up to 5 years.

The study was designed and conducted to adhere to Good Clinical Practice per the International Conference on Harmonisation Guideline E6 requirements and in accordance with the ethical principles outlined in the Declaration of Helsinki. The study was conducted in accordance with applicable national, state, and local laws per each study site’s Institutional Review Board, independent ethics committee, and pertinent regulatory authorities. All patients provided written informed consent before any study-related procedures were performed.

Study endpoints

The primary endpoint was PFS, as assessed by the independent review committee (IRC, i.e., central review) per 2007 International Working Group (IWG 2007) criteria [26], but without positron emission tomography imaging. Secondary endpoints included overall response rate (ORR), complete response (CR), duration of response (DOR), overall survival (OS), event-free survival (EFS), time to next anti-lymphoma treatment (TTNLT), and safety. Primary efficacy analyses were conducted in the intent-to-treat (ITT) population, defined as all patients randomized into the trial, regardless whether they received study treatment or not. All efficacy responses were assessed by a blinded, independent central review assessment of radiologic data using the 2007 IWG criteria, and progression or relapse was confirmed based on computed tomography/magnetic resonance imaging scans. Bone marrow biopsy was required to confirm a CR.

The safety analysis population was defined as all patients who received at least one dose of study treatment. Adverse events (AEs) were classified using the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE) version 4.03, except for tumor flare reaction, which was graded using the NCI CTCAE version 3.0, and tumor lysis (assessed per Cairo–Bishop Criteria) [27].

Statistical methods

The primary objective of AUGMENT was to show superior PFS in the R2 group compared with the R-placebo group, using a one-sided α = 0.025 level. The study was not designed with sufficient power to detect differences in an exploratory single-country subgroup analysis such as presented here. PFS and other time-to-event data were characterized by the Kaplan–Meier procedure [28]. A hazard ratio (HR) with a two-sided 95% CI was estimated using the stratified Cox proportional hazard regression model. For binary secondary efficacy endpoints, the number and percentage of patients were tabulated by treatment group, and a stratified Cochran–Mantel–Haenszel test to adjust for possible confounding effects of the stratification factors was performed. Prespecified subgroup analysis of PFS was also performed, with HR estimated from a Cox proportional hazard regression model.

Results

Patient population

From October 28, 2014, through June 6, 2016, a total of 36 Japanese patients were enrolled into the study, with 18 patients each in the R2 and R-placebo groups. Baseline characteristics were similar in the two groups (Table 1). Median age was 61 years (range 44–83). Thirty-five (97%) patients had FL and only 1 (3%) patient had MZL (R-placebo group, MALT subtype). Of the patients with FL, 25% had a high-risk Follicular Lymphoma International Prognostic Index score. Nine (25%; R2, n = 4; R-placebo, n = 5) patients had ≥ 3 prior systemic anti-lymphoma treatments. Relapse or progression within 2 years of their initial diagnosis had occurred in 13 (36%) patients, and 1 (3%) R2 patient was refractory to last regimen. The Japanese patient population was similar to the global AUGMENT population [25]; however, some differences (Japanese vs global) were noted, including Eastern Cooperative Oncology Group Performance Status 0 (92% vs 68%), high tumor burden (39% vs 51%), MZL histology (3% vs 18%), B symptoms (0% vs 8%), prior rituximab treatment (100% vs 84%), and refractory to last regimen (3% vs 16%).

Table 1 Baseline patient demographics and clinical characteristics of the Japanese and global populations (ITT population)
Full size table

Treatment completion and dose intensity

In the ITT population, the full treatment course was completed in 61% of patients in the R2 group and 83% of patients in the R-placebo group. In the safety population, the median number of lenalidomide/placebo treatment cycles was 12 (range, 4–12) for R2 and 12 (range 6–12) for R-placebo. The median relative dose intensity of rituximab was 96% in the R2 group and 98% in the R-placebo group. The median relative dose intensity of lenalidomide was 91% and the average daily dose was 17.30 mg (interquartile range, 4.90 mg from Q3–Q1 [20–15.10 mg]) in the R2 group.

In the R2 vs R-placebo groups, treatment discontinuations occurred in 39% vs 17% of patients, respectively. The most common reasons for discontinuation of lenalidomide or placebo were AE (17%) in the R2 group and PD (17%) in the R-placebo group. One patient in the R2 group withdrew from the study and discontinued rituximab. AEs leading to dose reduction/interruption of lenalidomide or placebo occurred in 44%/67% vs 11%/33% of patients in the R2 and R-placebo groups, respectively. AEs leading to dose interruption of rituximab occurred in 56% of R2 vs 50% of R-placebo patients. AEs leading to early discontinuation of lenalidomide or placebo occurred in 17% of R2 vs 0% of R-placebo patients.

In the R2 vs R-placebo group, the median time to first dose reduction of lenalidomide or placebo was 106 days (range, 12–288) vs 186 days (range, 78–294) and median time to first dose interruption was 58 days (range 6–261) vs 113 days (range, 29–304). Among patients in the R2 group, neutropenia was the most common reason for lenalidomide dose reduction (28%), interruption (44%), and discontinuation (11%).

Efficacy

The primary endpoint of PFS by IRC was superior in the R2 group vs the R-placebo group (HR 0.32; 95% CI 0.11–0.96). Median PFS as assessed by IRC in the R2 vs R-placebo group was not reached (NR) (95% CI 19.7–NE) vs 16.5 months (95% CI 11.3–30.6; Fig. 1). PFS as assessed by investigator also showed HR in favor of R2 (HR 0.62; 95% CI 0.23–0.1.67); median PFS by investigator was NR (95% CI 16.7–NE) vs 19.3 months (95% CI 13.9–NE) for the R2 and R-placebo groups, respectively.

Fig. 1
figure 1

Progression-free survival, as assessed by IRC in the ITT population

Full size image

Best ORR as assessed by IRC was higher in the R2 group, 94% (95% CI 73–100) compared with the R-placebo group, 56% (95% CI 31–79), with CR rates of 17% (95% CI 4–41) and 11% (95% CI 1–35), respectively (Table 2). Best response rates as assessed by investigator were also higher in the R2 vs R-placebo group. OS results were immature, with 1 death reported (R2 group; arrhythmia; Fig. 2).

Table 2 Efficacy outcomes (ITT population)
Full size table
Fig. 2
figure 2

Overall survival in the ITT population

Full size image

DOR (HR 0.4; 95% CI 0.13–1.25), EFS (HR 0.36; 95% CI 11.3–30.6), and TTNLT (HR 0.49; 95% CI 0.12–2.00) all had HRs in favor of R2. Median DOR, median EFS, and median TTNLT were not reached in the R2 group compared with 19.0 months, 16.5 months, and 37.1 months, respectively, in the R-placebo group (Supplementary Fig. 1–3). The median follow-up was 33.0 months.

Safety and tolerability

All 36 patients received treatment and were included in the safety population. Eighteen patients in the R2 group (100%) and 17 in the R-placebo group (94%) had AEs that occurred within 28 days after last dose (Table 3). Any-grade AEs that occurred more frequently (≥ 20% difference) in the R2 vs R-placebo group included infusion-related reaction (72% vs 44%), neutropenia (61% vs 33%), constipation (56% vs 11%), diarrhea (39% vs 0%), nasopharyngitis (39% vs 28%), rash (39% vs 17%), rash maculopapular (28% vs 0%), and increased alanine aminotransferase (28% vs 0%). More patients in the R2 group (67%) had at least one grade 3/4 AE compared with the R-placebo group (22%). Grade 3/4 AEs reported in > 1 patient in either group (R2 vs R-placebo) included neutropenia (50% vs 17%), decreased lymphocyte count (17% vs 11%), and decreased white blood cell count (6% vs 11%). One patient died on study (R2 group, arrhythmia, cycle 7, related to lenalidomide) and 1 patient had a second primary malignancy (SPM) (R2 group, invasive carcinoid tumor of the gastrointestinal tract). Time to onset of SPM was 19.6 months.

Table 3 All-grade (≥ 20% in either group) and grade ≥ 3 treatment-emergent adverse events (safety population)
Full size table

Discussion

R2 in Japanese patients from AUGMENT demonstrated superior efficacy compared with R-placebo and reduced the risk of progression by 68% (HR 0.32; 95% CI 0.11–0.96) compared with R-placebo. Median PFS was not reached in the R2 group compared with 16.5 months in the R-placebo group. Superior efficacy of R2 was also evident in the secondary endpoints of response rate, DOR, EFS, and TTNLT.

AEs were more common in the R2 group, largely due to higher rates of grade 3/4 neutropenia. Two patients in the R2 group discontinued lenalidomide due to neutropenia; however, neutropenia was predictable and manageable primarily through dose modifications and growth factor support. Growth factor support was used in 4 (22%) and 3 (17%) patients in the R2 and R-placebo groups, respectively. There were no instances of febrile neutropenia in either treatment arm.

The data for the Japanese subgroup were generally consistent with the global AUGMENT study [25]. There were no clinically relevant imbalances between the Japanese subgroup and the global population. R2 efficacy results were similar between the Japanese subgroup and global population, including median PFS HR (0.32 vs 0.46), PFS probability at 2 years (69% vs 58%), and ORR/CR (94%/17% vs 78%/34%). Additionally, tolerability of R2 was similar in the Japanese subgroup and global population. Any grade R2 associated AEs that occurred with a ≥ 25% difference between global vs Japanese patients included constipation (26% vs 56%), infusion-related reaction (15% vs 72%), and rash (11% vs 39%); however, R2 associated grade 3/4 AEs were similar between the Japanese and global populations, including rates of grade 3/4 neutropenia which occurred in 50% in both populations. In the R-placebo control arms, safety and efficacy data in Japanese patients were also similar to data in the global population. There was only one reported histological transformation in the Japanese population (R2 group).

Current Japanese guidelines for treating patients with relapsed iNHL, not refractory to rituximab, include rituximab monotherapy, bendamustine ± rituximab, fludarabine ± rituximab, multi-agent chemotherapy ± rituximab, radiotherapy, radioimmunotherapy, and other options; however, relative superiority of these options is unclear [9]. Similar to the global AUGMENT data, the data presented here demonstrated that R2 offered a clinically meaningful improvement compared with R-placebo in Japanese patients and had a tolerable safety profile. Although this was not a comparative study with other chemotherapies, these results suggest that R2 should be a new treatment option for Japanese patients with R/R iNHL.