Ibrutinib With Rituximab in First-Line Treatment of Older Patients With Mantle Cell Lymphoma

Preetesh Jain, MD, DM, PhD; Shuangtao Zhao, PhD; Hun Ju Lee, MD; Holly A. Hill, MPH; Chi Young Ok, MD; Rashmi Kanagal-Shamanna, MD; Fredrick B. Hagemeister, MD; Nathan Fowler, MD; Luis Fayad, MD; Yixin Yao, PhD; Yang Liu, PhD; Omar B. Moghrabi, BS; Lucy Navsaria, MBBS; Lei Feng, MS; Graciela M. Nogueras Gonzalez, MPH; Guofan Xu, MD; Selvi Thirumurthi, MD; David Santos, MD; Cezar Iliescu, MD; Guilin Tang, MD, PhD; L. Jeffrey Medeiros, MD; Francisco Vega, MD, PhD; Michelle Avellaneda, BS; Maria Badillo, BS; Christopher R. Flowers, MD; Linghua Wang, PhD; Michael L. Wang, MD


J Clin Oncol. 2022;40(2):202-212. 

In This Article


Patient Characteristics and Disposition

Fifty previously untreated elderly patients with MCL were enrolled in the study between October 2015 and November 2019. Baseline patient characteristics are summarized in Table 1. The median age was 71 years (interquartile range [IQR] 69–76 years), and 77% were men. Bone marrow involvement and GI tract involvement were observed in 94% (47 of 50) and 79% (19 of 24) of evaluable patients, respectively. High mantle cell lymphoma international prognostic index (MIPI) risk score was observed in 16%, and high biologic MIPI score including Ki-67% was observed in 28% of patients. TP53 aberrations were detected 3 of 18 (17%) patients in the bone marrow by either targeted next-generation sequencing (NGS) or fluorescence in situ hybridization, whereas 4 of 45 (8%) evaluable patients had complex karyotype. Eleven patients (23%) had a history of atrial fibrillation.

After a median follow-up of 45 months (IQR 24–56 months), five patients had died and 45 patients were alive. Overall, four patients developed disease progression (including three patients who transformed from classic to blastoid MCL in two and from classic to pleomorphic in one). Among these four patients with progression, the Ki-67% at baseline was 15%, 20%, 30%, and 35% and one patient had a TP53 mutation. Figure 1 shows the flowchart of patients.

Figure 1.

Flowchart of patient treatment and disposition. The induction treatment consisted of ibrutinib administered at 560 mg once daily on days 1–28 of a 28-day cycle and rituximab weekly for 4 weeks during cycle 1 and then day 1 of every cycle starting in cycles 3–8. After cycle 8, rituximab was given on day 1 of every 2 months for up to 2 years, and after 2 years, ibrutinib was administered in continuous cycles until disease progression or unacceptable toxicity or any other reason of discontinuation. Of the 50 patients enrolled, 28 came off study for various reasons.

As of January 2021, 28 of 50 (56%) patients discontinued IR therapy and came off study for various reasons (4 of 50 [8%] disease progression, 21 of 50 [42%] because of toxicities [including 10 because of grade 3 atrial fibrillation], and 3 of 50 [6%] for miscellaneous reasons). At the time of study discontinuation, 16 of 28 (57%) patients were in CR. Among the 10 patients with atrial fibrillation, six patients were new onset and four patients had a history of atrial fibrillation. Among the 14 patients who discontinued IR therapy for miscellaneous reasons, two were due to other cancers, three were due to bleeding, three were due to infections, five were due to intolerance, and one was by patient choice. The causes of discontinuation (n = 28) and causes of death (n = 5) are summarized in Table 2.


The investigator-assessed best ORR was 96% (46 of 48), CR 71% (34 of 48), PR 25% (12 of 48), and stable disease 4% (2 of 48). Two patients were not evaluable for response assessment since they came off study within 1 month after therapy initiation (one because of GI bleed and the other because of grade 3 atrial fibrillation).With an intent to treat, the best ORR was 92% (46 of 50) and CR was 68% (34 of 50). Forty patients had baseline PET-computed tomography scans performed, and 35 were positive. Among these 35 patients, 26 achieved complete metabolic response at best response (74%), eight did not have PET scan at best response, and one patient had residual disease. Of the 26 patients with CMR, 21 (81%) had bone marrow–negative for MCL, one had residual disease, and four patients did not have bone marrow evaluation. The median number of IR cycles to reach CR was seven (range 2–51).

We further evaluated responses in patients with Ki-67% (< 30%) and Ki-67% (≥ 30%-50%), and the n of N, ORR (n of N; CR) were 37 of 38, 97% (28 of 38; 74%) and 9 of 12, 75% (6 of 12; 50%), respectively (P < .001). Within the 18 evaluable patients for TP53 aberrations (three positive and 15 negative for TP53 aberrations), the n of N, ORR (CR) were 2 of 3, 66% (1 of 3, 33%) and 14 of 15, 94% (10 of 15, 67%), respectively, P < .001. Bone marrow flow cytometry assessment was performed at their best response in 32 patients, and 27 of 32 patients (84%) were negative for MCL cells.

Time to Event Outcomes

With a median follow-up of 45 months (IQR 24–56 months), overall, five patients had died and seven patients progressed or died (whichever occurred earlier). None of the deaths were on study. The median PFS and OS were not reached, and the 3-year PFS and OS were 87% (95% CI, 0.73 to 0.94) and 94% (95% CI, 0.82 to 0.98), respectively (Figs 2A and 2D). Patients with high Ki-67% (≥ 30%-50%) had a trend of higher risk of progression and/or death compared with those with low Ki-67% (< 30%), P = not significant (Figures 1E and 2B). Patients who achieved CR as their best response had significantly longer PFS and OS compared with those who did not achieve CR, Figures 1F and 2C.

Figure 2.

Survival outcomes after a median follow-up of 45 months. (A) The median PFS in all patients was not reached. (B) PFS by Ki-67% was not significantly different in high (Ki-67% ≥ 30%) versus low Ki-67% (< 30%) although higher hazard of progression was noted in those patients with high Ki-67%, HR of 2.62, P = .190. (C) PFS by CR status. Patients who achieved CR as the best response to IR therapy had a significantly better PFS compared with those patients without CR as the best response, P = .003. (D) The median OS in all patients was not reached. (E) OS by Ki-67% was not significantly different between low and high Ki-67% categories (P = .356). (F) OS by CR status. Patients who achieved CR as the best response to IR therapy had a significantly better OS compared with those patients without CR as the best response, P = .002. CR, complete response; HR, hazard ratio; IR, ibrutinib-rituximab; OS, overall survival; PFS, progression-free survival.

Furthermore, significantly higher risk of progression was noted in patients with high-risk simplified and modified MIPI score (Data Supplement). No difference in PFS was observed in patients with or without TP53 aberrations and those with or without complex karyotype (Data Supplement); however, OS was inferior in high-risk patients (Data Supplement).

Adverse Events

The adverse event profile on IR therapy is summarized in Table 3. Most adverse events were grade 1 or 2 (≥ 50% frequency—fatigue, neuropathy, diarrhea, myalgia, and oral mucositis were common). The most frequent grade 3–4 toxicities were 22% atrial fibrillation (n = 11), 18% fatigue (n = 9), 14% diarrhea (n = 6), and 14% myalgias (n = 7). Grade 3–4 hematologic toxicities were 4% anemia (n = 2), 8% neutropenia (n = 4), and 4% thrombocytopenia (n = 2). Four patients developed grade 3–4 bleeding while on ibrutinib (included hematuria, bronchopulmonary hemorrhage, GI bleeding, and retinal bleeding), 3 of 4 patients were on aspirin and/or enoxaparin (two on both and one only on aspirin), and one patient with retinal bleeding had recent glaucoma surgery and had history of vitreous surgery. One patient had grade 3 hematuria and Gram-negative urinary tract infection, which resolved after antibiotics and therefore resumed on study and maintained remission. None had grade 5 toxicities.

Overall, 17 of 50 (34%) patients developed atrial fibrillation and 10 of 17 had discontinued ibrutinib because of grade 3 atrial fibrillation. Of these 17 patients, nine patients (53%) were without a history of atrial fibrillation and six of them had baseline ECG abnormalities (including first-degree atrioventricular block in one, right bundle branch block in one, sinus bradycardia in three, and one with miscellaneous abnormalities). Eight patients had a history of atrial fibrillation. Fourteen patients were receiving 560 mg ibrutinib once daily at the onset of atrial fibrillation, whereas one patient each was taking ibrutinib 420, 280, and 140 mg once daily, respectively. The median age of patients who developed atrial fibrillation was 71 years (range 66–82 years), the median ejection fraction in pretreatment echocardiogram was 60% (range 44–75%), and the median time to onset of atrial fibrillation from the start of ibrutinib treatment was 9.5 months (range 1–48 months). In the Data Supplement, we have described the characteristics of patients who developed atrial fibrillation and those patients who did not develop atrial fibrillation. Of particular note, patients with atrial fibrillation had a higher median number of baseline cardiovascular risk factors, four (range 1–8), and by contrast, the median number of baseline cardiovascular risk factors in those patients who did not develop atrial fibrillation was two (range 1–4). Overall, dose reduction of ibrutinib was performed in 29 (58%) patients for various reasons (seven atrial fibrillation [two improved, whereas five patients had persistent atrial fibrillation or had recurrence], five infections, five bleeding, five myalgias, and seven miscellaneous).

Genomic Profiles in Patients according to the Response to IR Combination

We then performed WES in 25 pretreatment tissue biopsies. We divided the patients according to best response on IR therapy (CR and PR). Figure 3A depicts the pattern of somatic mutations and distribution according to CR versus PR. Patients with PR were enriched in KMT2D, FAT4, ROS1, CARD11, ATM, NOTCH1, CCND1, and FAT1 mutations. In Figure 3B, we evaluated the copy number variation and identified deletions in FAT1, FAT4, ROS1, and KMT2C were predominantly observed in patients with PR. Gain of MALT1 in the CR group and gain of SMARCA4 in the PR group were observed. Intratumoral heterogeneity and domains of specific proteins (KMT2D, CCND1, and NCOR2 mutants) are shown in the Data Supplement.

Figure 3.

Somatic mutation profile by WES and transcriptomic profile by bulk RNA sequencing and differential gene expression in patients, on the basis of achieving CR (C1) or PR (C2) after treatment with ibrutinib-rituximab. (A) The landscape of somatic mutations from pretreatment MCL samples (n = 25). The bottom panel shows somatic mutations and gene-level copy number alterations by sample (column) and by gene (row). The middle tracks display the clinical characteristics. The histogram on the top shows the number of alterations accumulated on 28 listed genes in each individual patient. The right bar plots show the composite of all mutations between CR (C1) and PR (C2) groups. Fisher's exact test, P < .05. (B) Composite of copy number profiles between CR and PR groups, with gains in red and losses in blue. The regions that showed a difference in the frequency of copy number alterations between two subtypes are shaded in light red rectangles, and within which the names of cancer-related or biologically important genes are labeled (STAB1, FAT1, FAT4, KMT2C, MALT1, SMARCA4, ROS1, NCOR2, and RB1). (C) Transcriptomic profile of baseline tumor specimens from 16 patients with MCL is shown, CR (C1) and PR (C2). Unsupervised hierarchical clustering of DEGs on RNA-seq analysis is shown. Genes with log2(fold change) > 1 and a FDR q < 0.05 were applied to filter DEGs. Biologically important genes are labeled on the right of the plot. The top tracks show clinical characteristics among the samples. A Fisher's exact test is used to identify significant clinical factors; response was significantly correlated with the DEGs (P = .003). (D) Violin plot indicates biologically important DEGs among the three clusters. BM, bone marrow; CR, complete response; DEG, differentially expressed gene; FDR, false discovery rate; MCL, mantle cell lymphoma; PR, partial response; WES, whole-exome sequencing.

Bulk RNA Sequencing and DEG According to the Response to IR Combination. Sixteen baseline samples were evaluable for bulk RNA sequencing. A distinct pattern of differentially expressed genes including CCND1, BIRC3, BANK1, SETBP1, RRAS2, AXIN2, and IL2RA was observed in patients who achieved a PR (n = 7) compared with patients who achieved CR (n = 9; Figs 3C and 3D). In the Data Supplement, pathways that are differentially enriched in PR versus CR are depicted. Gene set enrichment pathway analysis demonstrated that B-cell receptor (BCR) pathways were predominantly upregulated in PR patients (Data Supplement), whereas mitogen-activated protein kinase, TP53, calcium signaling, and hypoxia-associated pathways were upregulated in those with CR.