Unmet needs in relapsed/refractory mantle cell lymphoma (r/r MCL) post-covalent Bruton tyrosine kinase inhibitor (BTKi): a systematic literature review and meta-analysis

Abstract

To quantify the clinical unmet need of r/r MCL patients who progress on a covalent Bruton tyrosine kinase inhibitor (BTKi), we conducted a systematic review to identify studies that reported overall survival (OS), progression-free survival (PFS), or response outcomes of patients who received a chemo(immunotherapy) ± targeted agent standard therapy (STx) or brexucabtagene autoleucel (brexu-cel) in the post-BTKi setting. Twenty-six studies (23 observational; three trials) reporting outcomes from 2005 to 2022 were included. Using two-stage frequentist meta-analyses, the estimated median PFS/OS for patients treated with an STx was 7.6 months (95% CI: 3.9–14.6) and 9.1 months (95% CI: 7.3–11.3), respectively. The estimated objective response rate (ORR) was 45% (95% CI: 34–57%). For patients treated with brexu-cel, the estimated median PFS/OS was 14.9 months (95% CI: 10.5–21.0) and 32.1 months (95% CI: 25.2–41.2), with a pooled ORR of 89% (95% CI: 86–91%). Our findings highlight a significant unmet need for patients whose disease progresses on a covalent BTKi.

Keywords:

• Mantle cell lymphoma
• unmet clinical need
• systematic literature review
• meta-analysis
• standard of care
• CAR T-cell therapy

Introduction

Mantle cell lymphoma (MCL) is a rare, aggressive type of B-cell non-Hodgkin lymphoma (NHL) that is typically diagnosed in patients aged over 65 years and more often diagnosed in men [1]. Despite advances in diagnostic technologies to reduce the number of false positives, the incidence of MCL in the United States has continued to trend upward since the 1990s and epidemiological estimates suggest that MCL accounts for 2–6% of all NHL diagnoses [2–4]. It is heterogeneously characterized with a range of variants, relevant biological indicators and, consequently, varied treatment pathways [5].

Front-line treatment is typically comprised of chemoimmunotherapy regimens, which have historically been followed by rituximab maintenance [5,6]. The course of treatment is typically risk-stratified according to the mantle cell international prognostic index (MIPI), a composite measure which factors in a patient’s age, performance status, lactase dehydrogenase (LDH), and leukocyte count [7]. Other important biological qualifiers have been identified for MCL, such as the cell proliferation rate, measured as Ki-67, and the presence of a TP53 gene mutation, which are both considered reliable prognostic factors in predicting survival outcomes [5,6,8].

Despite the typically aggressive front-line clinical course which results in adequate responses, patients whose treatments fail historically face a bleak prognosis [7,9]. The introduction of covalent Bruton tyrosine kinase inhibitors (BTKis), marked by the 2013 approval of ibrutinib for the second-line treatment of MCL, has improved clinical outcomes among high-risk patients who progress following front-line therapy [10]. The approval for ibrutinib cited a pivotal single-arm trial (NCT01236391) and its results have been further supported by additional studies among patients with intermediate-to-high risk relapsed/refractory (r/r) MCL [11]. Novel BTKis have emerged since, including acalabrutinib (second-generation BTKi) and zanubrutinib, which have demonstrated favorable efficacy and safety profiles [12,13]. Yet, treatment with BTKi is not without complications and, while treatment may yield initial success, long-term BTKi use, such as with ibrutinib, is characterized by difficulties in maintaining long-term remission, development of treatment resistance, and reports of off-target effects such as cardiac arrhythmias [10,14,15].

While the development of covalent BTKis has improved the outcomes of patients who progress following front-line therapy, the issue of a poor prognosis continues to be a challenge for patients with r/r MCL who progress on a BTKi. For these patients, most chemotherapy salvage treatments are ineffective and durations of response, if achieved, are short [14,16]. Although this is well understood, the unmet clinical need for these patients is poorly quantified in the literature. As the use of BTKis has increased, studies which quantify this potential unmet need have become increasingly important to support clinicians and decision makers in understanding how emerging treatments are addressing this gap.

Until recently, impactful and reliable treatments for these patients have been elusive. The chimeric antigen receptor (CAR) T-cell therapy brexucabtagene autoleucel (brexu-cel) has been heralded as a breakthrough in this clinical space, with strong outcomes in the international, single-arm ZUMA-2 clinical trial [17]. The success of this trial led to the approval of brexu-cel for adult patients with r/r MCL who had previously been treated with two lines of therapy, including a BTKi, by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) in 2020 [18,19]. In addition to improving survival, brexu-cel appears to lead to a durable response. Despite the encouraging results, some researchers have expressed concern over the size and duration of the trial [20]. To this end, a growing literature of the real-world use of brexu-cel in r/r MCL could help strengthen the conclusions of ZUMA-2.

The objective of this systematic review and meta-analysis was to characterize and quantify the unmet need in patients treated with a chemo(immunotherapy) ± targeted agent standard therapy (STx). Secondarily, to contrast these outcomes, studies of patients treated with brexu-cel in this setting were similarly identified and synthesized.

Methods

This systematic literature review and meta-analysis was conducted according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [21].

Literature search

Systematic searches of MEDLINE, EMBASE, and Evidence-Based Medicine Reviews (EBM Reviews) were conducted on 26 October 2022. These searches were supplemented by hand searches of the conference proceedings from the 2022 American Society of Hematology (ASH) annual meeting which became available after the database search date but were anticipated to include several publications of relevance and importance. Study eligibility was determined using pre-defined PICOS (population, intervention, comparator, outcomes, and study design) criteria. Eligible studies included adults with r/r MCL who were receiving a post-BTKi STx (defined as chemo(immunotherapy) ± a targeted agent) or brexu-cel following treatment with a BTKi. To be considered a post-BTKi study, at least 80% of the patient cohort must have had a prior BTKi exposure [22]. The list of eligible interventions was extended to include any non-experimental systemic therapy, with or without concomitant interventions such as radiotherapy. All prospective and retrospective designs, including both observational studies and trials, were eligible for inclusion. Our motivation for including both clinical trial designs and real-world data was that while clinical trials can provide insights into the efficacy and safety of treatments, real-world studies provide further insights into unmet needs by reflecting outcomes in routine clinical practice. The outcomes of interest included three time-to-event outcomes – overall survival (OS), progression-free survival (PFS), duration of response (DOR) – and three dichotomous outcomes – objective response rates (ORRs), complete response (CR), and partial response (PR). Study eligibility was restricted to records published in English. As this systematic literature review was designed as an extension and expansion to a previously published review and meta-analysis [23], the search syntax was restricted to identify records published from November 2019 onward. A detailed search strategy is provided in Table S1.

Study selection and data extraction

All study selection and data extraction steps were conducted by two reviewers working independently and in duplicate in accordance with PRISMA guidelines [24]. Titles and abstracts, followed by the full-text publications of included abstracts, were evaluated against the review’s eligibility criteria. In situations where consensus could not be reached between the two reviewers, a third reviewer provided arbitration. Following the identification of eligible publications, related publications reporting on common patient cohorts were linked or ‘mapped’ prior to data extraction to avoid double-counting of outcomes in the planned meta-analyses. These linkages were established based on characteristics such as references to a common study protocol, institution names, sample sizes, and interventions.

Data extraction of study characteristics, patient characteristics, and outcomes was completed at the study-level using standardized, piloted data extraction templates from all mapped publications. Where published Kaplan–Meier (KM) curves were available, curve data was digitized using WebPlotDigitizer-4.5 (https://automeris.io/WebPlotDigitizer) and the corresponding numbers at risk were recorded for reconstruction of pseudo-individual patient data (IPD) according to the Guyot algorithm [25].

The quality of studies was assessed according to the Newcastle-Ottawa Scale (NOS), which is appropriate for observational prospective and retrospective studies as well as single-arm, non-comparative trials. The use of the NOS facilitated comparisons of methodological rigor across the evidence space as, while some studies may be initiated as randomized trials of BTKi, these also effectively become observational once patients progress to the post-BTKi setting. All data extraction and critical appraisal steps were completed independently and in duplicate by two reviewers. Discrepancies were resolved by discussion or, if a consensus could not be reached, through arbitration provided by a third reviewer.

Statistical analyses

Meta-analyses were used to estimate survival curves and response proportions. Whereas most meta-analyses aim to estimate treatment effects (i.e. to compare treatments), our aim was to estimate the clinical outcomes among r/r MCL patients receiving a post-BTKi STx – or receiving brexu-cel in the secondary analyses. Time-to-event outcomes (OS, PFS, and DOR) were analyzed using multivariate meta-analysis of survival function parameters as developed by Cope et al. [26]. For each survival curve, the pseudo-IPD were analyzed using alternative survival distributions – namely Weibull, log-normal, log-logistic, Gompertz, and exponential distributions. For each distribution, the survival parameters were meta-analyzed to fit an OS curve for the collection of curves. The goodness-of-fit was statistically assessed according to the Akaike information criteria (AIC) and supplemented by visual inspection. Uncertainty about the point estimates was described with the 95% confidence interval (CI) based on bootstrapping with 1000 samples. While all studies reporting survival outcomes were included in the review and narrative summary, only studies with available KM curves were included in the meta-analysis. From the pooled survival distributions, estimates of median survival as well as survival proportions at 12 and 24 months were made.

Dichotomous response outcomes (ORR, CR, and PR) were meta-analyzed to generate a point estimate pooled proportion with 95% CI. Heterogeneity in the evidence for each outcome was assessed according to τ and further qualified by the I² statistic [27]. All meta-analyses were conducted under a frequentist framework and included evaluations of both random and fixed effects models. All analyses were performed using R version 4.1.0 (R Foundation for Statistical Computing, Vienna, Austria).

Results

Evidence base

The systematic searches identified 1460 publications from the bibliographic databases, with 37 records satisfying all eligibility criteria. The evidence selection process is summarized in Figure 1. A further five publications were identified through conference hand searches and supplemental searches. Ten publications from an earlier iteration of this review were also included [23]. In total, 52 publications describing 26 studies were included in the evidence base, with 19 and 7 studies describing outcomes of patients treated with STx and brexu-cel, respectively (Table 1).

Figure 1. PRISMA flow of information for study selection. *Search dates purposefully overlapped with the previous iteration of this literature review.

Table 1. Studies of chemo(immunotherapy) ± targeted agent standard therapy and brexucabtagene autoleucel in the post-BTKi setting.

Study	Intervention	Study initiation context	Sample size		Geography	Study start date, study end date	Age, median (min, max)	Males, n (%)	BTKi exposure, n (%)	Median follow-up, months
			Start of BTKi	Post-BTKi
Chemo(immunotherapy) ± targeted agent standard therapy
Global 15 [28]	Rituximab, lenalidomide, cytarabine, bendamustine, bortezomib, anthracycline, PI3K inhibitor	BTKi	114	73	United States, United Kingdom, Germany, and Poland	–	67 (47, 85)	85/114 (75)	All exposed	–
Japan RWE [29,30]	BR, bendamustine, rituximab, R-CHOP, VR-CAP, BR-CHOP, R-HyperCVAD, and others	BTKi	247	137	Japan	April 2008–July 2021	77 (43, 95)	183/247 (74)	All exposed	–
Korea RWE [31]	Bendamustine, cytarabine, bortezomib, doxorubicin, fludarabine, etoposide	BTKi	88	16	Korea	January 2016–November 2019	71^a (42, 92)	71/88 (81)	All exposed	30.5
LYSA [32]	Rituximab, bendamustine, bortezomib, and dexamethasone (RiBVD)	BTKi	67	12	France	June 2016–January 2019	69^a (40, 91)	–	All exposed	–
MCL-004 [33]	Lenalidomide, lenalidomide + combinations	Standard therapy	–	58	United States, United Kingdom	March 2009–April 2016	71 (50, 89)	44/58 (76)	All exposed	–
MCL-3001 [34]	Rituximab, bendamustine, and cyclophosphamide	BTKi	139	44	EU, LATAM, Asia-Pacific	December 2012–November 2013	67^a	100/139 (72)	All exposed	20
MD Anderson – BCL Moon Shot [35–37]	Systemic therapy and chemo-immunotherapy	BTKi	21	15	United States	–	66 (43, 88)	–	All exposed	38
MD Anderson – RWE [14,38]	HyperCVAD, radiochemotherapy, bendamustine, lenalidomide, bortezomib, R-CHOP, radiation alone, R-ESHAP with alloSCT, lenalidomide + rituximab + proteasome inhibitor, and others	BTKi	80	41	United States	February 2011–March 2017	69 (35, 86)	29/34 (85)	All exposed	38
MD Anderson – Venetoclax cohort [39,40]	Venetoclax	Standard therapy	–	24	United States	–	69 (58, 82)	20/24 (83)	22/24 (92)	17.5
NPP RWE (Janssen-Cilag Named Patient Programme) [41,42]	Supportive care, low-intensity immunochemotherapy/radiotherapy, lenalidomide, or venetoclax	BTKi	65	16	United Kingdom, Ireland	November 2014–November 2019	67^a (48, 90)	49/65^a (76)	All exposed	60
Oxford University [43]	Venetoclax	Standard therapy	–	20	United Kingdom	March 2016–May 2018	69 (43, 84)	17/20 (85)	All exposed	–
RTL 10 [44]	Bortezomib, bortezomib + rituximab or bendamustine, lenalidomide, R-BAC, BR, Carfilzomib, lenalidomide, dexamethasone, venetoclax, rituximab and chlorambucil, HR-cytarabine, R-COMP	BTKi	69	14	Italy	2005–2019	70^a (41, 89)	45/69^a (65)	All exposed	15.6
SCHOLAR-2 [45]	R-BAC, BR, cytarabine, gemcitabine, BTKi, lenalidomide, bortezomib, and others	Standard therapy	–	149	United Kingdom, France, Germany, Spain, Italy, Sweden, Denmark	July 2012–July 2018	71 (43, 91)	108/149 (72)	All exposed	27.3
SEER [46]	Cytarabine, bendamustine, and anthracycline; lenalidomide, venetoclax, other chemoimmunotherapy or rituximab single agent and others	BTKi	334	100	United States	2007–2017	74^b (71, 79)	223/332^a (67)	All exposed	22.2
UK RWE – SoC [47–51]	R-BAC, R-CHOP, BR, and chemotherapy	BTKi	211	42	United Kingdom	October 2015–Jun 2020	70 (44, 85)	147/211^a (70)	All exposed	13
US 12 – SoC [52]	Venetoclax monotherapy, venetoclax + mAb, venetoclax + BTKi	Standard therapy	–	81	United States	–	69 (38, 88)	64/81 (79)	74/81 (91)	–
US 8 [53]	Bortezomib, lenalidomide, and bendamustine	BTKi	97	29	United States	November 2013–December 2015	63^c (46, 81)	24/34^c (71)	All exposed	–
US-Japan RWE (US only) [54,55]	Rituximab, BTKi based regimens, B-cell lymphoma 2 inhibitor, and chemotherapy	Standard therapy	–	352	Japan, United States	December 2010–October 2020	72^a	264/352 (75)	All exposed	–
VALERIA [56]	Venetoclax, lenalidomide, rituximab	Standard therapy	–	15	Finland, Denmark, Sweden, Norway	–	72^a (54, 81)	48/59^a (81)	All exposed	21
Brexucabtagene autoleucel
CIBMTR [57]	Brexu-cel	Post-BTKi	–	135	United States, Canada	July 2020–December 2021	66	107/135 (79)	120/132 (89)	6
DESCAR-T [58]	Brexu-cel	Post-BTKi	–	47	France	January 2020–June 2021	67 (45, 79)	44/47 (94)	All exposed	3.3
European Early Access Program [59,60]	Brexu-cel	Post-BTKi	–	39	Spain, Italy, Germany, and the Netherlands	February 2020–August 2021	67 (47, 79)	29/33 (88)	All exposed	10.1
UK RWE – Brexu-cel [19]	Brexu-cel	Post-BTKi	–	71	United Kingdom	July 2020–June 2021	67 (46, 80)	59/82 (72)	All exposed	7.4
US 12 – Brexu-cel [61]	Brexu-cel	Post-BTKi	–	52	United States	August 2020–December 2021	66 (47, 79)	43/52 (82)	All exposed	4.2
US Consortium [62–65]	Brexu-cel	Post-BTKi	–	189	United States	October 2016–July 2021	67 (34, 89)	128/168 (76)	144/168 (86)	14.3
ZUMA-2 [17,66–75]	Brexu-cel	Post-BTKi	–	74	United States	February 2021–June 2022	65 (38, 79)	–	All exposed	35.6

BR: bendamustine and rituximab; R-CHOP: rituximab, cyclophosphamide, doxorubicin hydrochloride, vincristine sulfate, and prednisone; VR-CAP: bortezomib and rituximab–cyclophosphamide, epirubicin and prednisone; BR-CHOP: bendamustine–rituximab, cyclophosphamide, doxorubicin hydrochloride, vincristine sulfate, and prednisone; R-HyperCVAD: rituximab, cyclophosphamide–vincristine–doxorubicin–dexamethasone–methotrexate–leucovorin–cytarabine; HyperCVAD: cyclophosphamide–vincristine–doxorubicin–dexamethasone–methotrexate–leucovorin–cytarabine; R-ESHAP: rituximab, etoposide, solu-medrone, high dose cytarabine, and cisplatin; R-BAC: rituximab, bendamustine, and cytarabine; R-COMP: rituximab, cyclophosphamide, non-pegylated liposomal doxorubicin, vincristine sulfate, and prednisone.

^a At the start of BTKI.

^b IQR.

^c Number based on ibrutinib non-responders.

Across the evidence base, most studies (17/26) were based at centers in North America. Two large international trials, ZUMA-2 [66] and MCL-3001 [34], were also included and recruited patients primarily from South America, North America, and Europe. Study protocols were initiated at different times, with 12 studies enrolling patients for BTKi administration and reporting downstream outcomes post-BTKi and 14 studies initiated in the post-BTKi setting. Most studies were of a retrospective observational design (23/26), with two (MCL-3001 [34]; VALERIA [56]) and one (ZUMA-2 [66]) single-arm trials in the STx and brexu-cel settings, respectively. All studies of brexu-cel were initiated post-BTKi. Study outcome availability is summarized in Table S2.

Critical appraisal of the included studies did not reveal any methodological concerns that would distort outcome signals (Table S3).

Chemo(immunotherapy) ± targeted agents

For the primary analyses, outcomes from chemo(immunotherapy) ± targeted agent standard therapies were reported in 19 studies (N = 1238). A heterogeneous mix of interventions was reflected in the evidence base. Studies were comprised of cohorts treated with a common intervention as well as cohorts comprised of patients receiving one of several systemic treatments, including chemotherapy, immunotherapy, combination chemotherapy with immunotherapy, and with or without radiotherapy.

Fourteen studies (N = 1080) reported OS outcomes for post-BTKi standard therapies, where the median OS varied from 2.5 to 19.4 months. A log-normal distribution was fit to the published survival curves from 10 studies (n = 669), leading to an estimated median OS of 9.1 months (95% CI: 7.3–11.3), as illustrated in Figure 2(A). Survival estimates at 12 and 24 months were 42% and 25%, respectively (Table 2).

Figure 2. Meta-analysis of Kaplan–Meier curves from reconstructed individual patient data for overall survival and progression-free survival. (A) Meta-analysis of overall survival outcomes from studies of chemo(immunotherapy) ± targeted agent standard therapy, summary curve based on a log-normal distribution; (B) meta-analysis of progression-free survival outcomes from studies of chemo(immunotherapy) ± targeted agent standard therapy, summary curve based on an exponential distribution; (C) meta-analysis of overall survival outcomes from studies of brexucabtagene autoleucel, summary curve based on an exponential distribution; (D) meta-analysis of progression-free survival outcomes from studies of brexucabtagene autoleucel, summary curve based on a log-normal distribution.

Table 2. Summary of available evidence and meta-analyses.

Outcome	Measure	Brexucabtagene autoleucel			Chemo(immunotherapy) ± targeted agent standard therapy
		Number of studies^b	Pooled sample size	Estimate (95% CI)	Number of studies^b	Pooled sample size	Estimate (95% CI)
Overall survival^a	Median survival, months	5	371	32.11 (25.20, 41.23)	10	669	9.07 (7.26, 11.28)
	Percent alive at 12 months			77%			42%
	Percent alive at 24 months			60%			25%
Progression-free survival^a	Median survival, months	6	418	14.90 (10.53, 20.95)	5	80	7.58 (3.86, 14.61)
	Percent without PFS at 12 months			56%			33%
	Percent without PFS at 24 months			37%			11%
Objective response rate	Proportion with response	7	546	0.89 (0.86, 0.91)	13	414	0.45 (0.34, 0.57)
Complete response	Proportion with response	7	546	0.74 (0.68, 0.80)	11	360	0.23 (0.14, 0.34)
Partial response	Proportion with response	7	360	0.14 (0.09, 0.20)	11	360	0.23 (0.16, 0.30)
Duration of response	Median duration, months	3	213	22.3 (16.09, 33.01)	1	9	8.1
	Percent with response at 12 months			64%			–
	Percent with response at 24 months			48%			–

^a Overall survival and progression-free survival meta-analysis outcomes presented based on the best-fitting survival extrapolation model.

^b Number of studies included in the meta-analysis.

Progression-free survival was sparsely reported in the STx setting during the time period under review. Across the five studies (n = 169) where median PFS was reported, estimates varied from 1.9 to 10.1 months. Published KM curves were available from three studies (n = 80) and an exponential distribution was used to estimate a pooled median PFS of 7.6 months (95% CI: 3.9–14.6) as visualized in Figure 2(B). The estimated PFS at 12 and 24 months was 33% and 11%, respectively (Table 2). Meta-analyses of OS and PFS were not sensitive to the selection of model distribution.

Thirteen studies (n = 414) were included in the meta-analysis of ORR, with a random effects pooled percentage of 45.3% (95% CI: 33.9–56.9%) as illustrated in Figure 3(A). The pooled estimate was characterized by a significant level of heterogeneity (I²: 80%, p < .01). Eleven studies (n = 360) were included in meta-analyses of CR and PR, with 23.1% (95% CI: 13.9–33.6%) and 22.8% (95% CI: 16.3–30.0%) of patients achieving a complete or PR, respectively (Table 2). High heterogeneity was also observed in the analyses of both CR and PR, similar to the meta-analysis of ORR, and reflective of differences in standard therapies, treatment histories, and clinical settings. Among responders to a STx, the DOR was similar to PFS. Six studies reported DOR outcomes, and a KM survival curve was only available in a single publication [43]. In this study, the 9/17 patients who achieved a response with venetoclax maintained response for a median 8.1 months (95% CI: 2.8–9.8).

Figure 3. Objective response rate, by treatment category. (A) Patients treated with chemo(immunotherapy) ± targeted agent standard therapy. (B) Patients treated with brexucabtagene autoleucel.

Brexucabtagene autoleucel

For the secondary analyses, outcomes for patients treated with brexu-cel were available from seven studies (N = 607). While most studies enrolled only BTKi-exposed patients, two retrospective studies enrolled mixed cohorts with 86% and 89% of patients treated with a BTKi [57,62]. Most studies (4/7) were conducted in North America, with others enrolling patients at European sites. Primary analyses are based on outcomes for infused cohorts, as these were more consistently reported in the literature, with secondary analyses based on intent-to-treat populations.

Seven studies reported brexu-cel OS outcomes (N = 546), with observed 6-month survival estimates varying from 79% to 93%. The pooled median OS, when an exponential distribution was fit to the published survival curves from five studies, was 32.1 months (95% CI: 25.2–41.2; n = 371) as illustrated in Figure 2(C). Survival estimates at 12 and 24 months were 77% and 60%, respectively (Table 2). Progression-free survival was available in all seven studies (N = 470), with observed 6-month PFS estimates varying from 58% to 83%. Based on the six studies with published KM curves (n = 418), a log-normal distribution was used to estimate a pooled median PFS of 14.9 months (95% CI: 10.5–20.9), as illustrated in Figure 2(D). Estimates of PFS at 12 and 24 months were 56% and 37%, respectively (Table 2). Across all analyses, the estimates for OS and PFS were not sensitive to the selection of the survival model. Analyses of OS and PFS based on intent-to-treat populations, referring to patients who underwent leukapheresis, were feasible with data from three studies (N = 299) and generated similar estimates of effect (Table S4).

Response outcomes (ORR, CR, and PR) were available from all seven studies of brexu-cel (N = 546; Figure 3(B)). The pooled ORR was 88.6% (95% CI: 85.7–91.2%) based on a random effects model and characterized by an absence of heterogeneity (I²: 0%) which is indicative of a strong and consistent signal of effect. This ORR estimate was primarily driven by CR, with a pooled rate of 74% (95% CI: 68–80%; I²: 68%, p = .03). Duration of response KM curves was available from two studies, with a pooled median estimate of 22.3 months (95% CI: 16.1–30.1; n = 213) based on a log-normal distribution. An estimated 64% and 48% of patients maintained response at 12 and 24 months, respectively.

Discussion

Our systematic literature review and meta-analysis quantified and confirmed a clear and persistent unmet clinical need. Treatment options have historically been limited for patients in the r/r MCL setting. For patients treated with a chemo(immunotherapy) ± targeted agent STx, median survival estimates were consistently short and fewer than half of patients responded to treatment. These observations are based on the collective findings of several studies, which together reflect the heterogeneous and diverse clinical nature of this setting. Not surprisingly, these up-to-date findings echo that of earlier research in this field that has maintained urgent calls for more effective treatments.

The emergence of BTKi therapies has considerably improved patient outcomes among r/r MCL patients and provides a genuine second-line option. Indeed, clinical guidelines for the treatment of MCL, including guidelines by the National Comprehensive Cancer Network (NCCN), list BTKi as a preferred second-line treatment for advanced MCL [76]. Since the introduction of ibrutinib, further developments have been made, with the approval of second-generation and next-generation BTKis [77]. Second-generation BTKi therapies provide less off-target activity, which translates to fewer adverse events, while next-generation BTKis have been shown to be effective across high-risk groups. Nonetheless, numerous studies have suggested that patients who progress while receiving BTKi therapy have diminished outcomes, culminating in short survival. The aforementioned NCCN guidelines do not provide a preferred third-line treatment, but do acknowledge the use of brexu-cel for this setting [76]. Our study provides a definitive, contemporary overview of the unmet need for these patients, by identifying and synthesizing all available evidence. While the studies were heterogeneous, and the source of this heterogeneity was not immediately clear, they were consistent in demonstrating an unmet clinical need. Our secondary analyses also demonstrated the potential role that brexu-cel can play in addressing this need.

The promising clinical outcomes of patients treated with brexu-cel mark a significant improvement over historically poor clinical outcomes in this therapeutic space. The results of our systematic literature review show that there is a growing evidence base that supports the strong improvements in OS and DOR observed in patients treated with brexu-cel. Not only does this review offer a much larger sample size relative to ZUMA-2, but the results were consistent and outcomes in the real-world settings aligned well with those reported in the trial setting. This reinforces the clinical utility of brexu-cel across the broader clinical setting found in the real-world where patients have different comorbidities, prognostic and risk factors, and treatment histories, as well as responses to those previous treatments. To this end, clinicians managing patients in the r/r MCL setting often weigh the value of clinical predictors in decision-making. Several host factors, disease characteristics, and pathobiologic factors that may be prognostic for an increased risk in non-response or a loss-of-response to BTKi have been identified [78]. For instance, recent literature exploring prognostic factors following ibrutinib monotherapy found that many patients were considered high-risk based on the simplified MIPI (s-MIPI), blastoid morphology, and/or a TP53 mutation [79,80]. Similarly, patients with a Ki-67 index greater than 30% or 50% may have inferior response outcomes [8,47,81]. Further research is needed to identify and understand prognostic factors in the post-BTKi setting, particularly when considering the use of CAR T-cell therapy such as brexu-cel. Some evidence suggests a reduced effect of CAR T-cell therapy on some outcomes among patients with a TP53 mutation [62]. However, other emerging evidence, such as from the ZUMA-2 trial of brexu-cel in r/r MCL, suggests that the impact of many typical front-line prognostic factors, such as those listed above, may not carry as much weight in the context of cellular therapy [17,82]. These large, robust improvements in clinical outcomes mirror those shown by CAR T-cell therapies for other lymphoma indications, such as large B-cell lymphoma [83,84].

Looking forward, we can expect a shift in the post-BTKi population. In the current evidence base, patients treated with ibrutinib and other BTKi have primarily been second-line or higher, meaning that patients in the post-BTKi setting have primarily been in the third- or later line of therapy setting. Recently, first-line approaches with fixed-duration ibrutinib have been published [85,86]. This shift toward first-line use opens the possibility of other solutions to the unmet need. For example, in later relapses after a limited duration treatment, a BTKi re-exposure may be reasonable. However, whether the other salvage treatments (CAR T-cells vs. conventional) display different efficacy, as shown in the current meta-analysis, remains to be seen. Other treatment alternatives, such as allogeneic stem cell transplantation, have demonstrated durable remission for patients in post-BTKi and high-risk feature settings [87,88].

This literature review was designed with several methodological considerations to strengthen the confidence and credibility of its findings. In establishing the scale of the evidence base, deliberate efforts were made to identify studies which may report the outcomes of patients post-BTKi but may be missed through more traditional search and screening strategies. That is, the search strategy was designed primarily around BTKi terminology, with an understanding that investigators in this space would speak to patient treatment exposure in defining their clinical questions. Furthermore, publication of studies which enrolled patients for treatment with a BTKi was eligible for inclusion during the title and abstract screening phase so the full-text publication could be reviewed for post-BTKi outcomes. This approach led to the identification of six publications which would have been otherwise excluded. Several analytical approaches were adopted, including the consideration of multiple survival extrapolation models which were evaluated based on robust criteria. Finally, study credibility was assessed according to the NOS. This instrument was chosen as it is well-suited for evaluating both observational studies as well as non-comparative single-arm trials, making it an ideal tool for this clinical context. Evaluations with the NOS returned few points of methodological concern in the included studies.

While the findings and conclusions of this review are based on robust methodology, there are limitations to consider. Observational studies hold an inherent risk of selection bias in how patients are assigned to treatments, and, in the case of the current review, post-BTKi treatment protocols were not disclosed in the publications. This is primarily applicable to studies which initiate in the BTKi space and then report post-BTKi treatments and outcomes. Similarly, patients who continue to receive treatment following BTKi discontinuation may be healthier and more fit as they are able to continue with further lines of therapy. Despite this supposed clinical advantage, however, studies of chemo(immunotherapy) ± targeted agent STx repeatedly demonstrated failures in treatments achieving their intended effects. It is possible that this represents a selection bias for patients who were recommended to and received CAR T-cell therapy, as these patients may be more clinically fit for this treatment. Similarly, era effects may distort efficacy and effectiveness signals for the chemo(immunotherapy) ± targeted agents STx treatments which were collected over a longer time horizon. Indeed, there have been multiple advances in both active treatment options for earlier lines of therapy as well as for diagnostics and supportive care which contribute to better outcomes today than in the past [89]. Time-to-event analyses were restricted to studies where a published KM curve was available, thus restricting the number of studies which contributed to the meta-analysis. In meta-analyses of KM survival curves, a single survival extrapolation model must be selected to reflect all evidence inputs and, while the chosen model may be the best fit for most studies, it is possible it would not be the best fit for each individual study.

The treatment landscape for post-BTKi r/r MCL is prominently defined by an urgent unmet clinical need, with treatments that fail to generate responses and diminished survival expectations. A growing body of evidence consistently reflects the promising impact of brexu-cel, with high response and duration rates and extended survival. Its real-world uptake and study outside of clinical trials further reinforces the positive impact this novel treatment has had across a more diverse and clinically heterogeneous population. Further research, particularly outside of clinical trial settings, will further the profile for brexu-cel as an answer to the repeated calls for innovative and impactful treatments in the post-BTKi setting.

Author contributions

All authors had full access to all data in the study. All authors take responsibility for the integrity of the data, the accuracy of the data analysis, and the final decision to submit for publication. Study concept and design: all authors. Acquisition, analysis, or interpretation of data: JW, SK, and MJZ. Drafting of the manuscript: all authors. Critical revision of the manuscript for important intellectual content: all authors. Administrative, technical, or material support: JW, TI, and MW. Study supervision: MD, BS, and MW.

Acknowledgements

David Wennersbusch and Fang Zhu assisted with some of the systematic literature review steps.

Disclosure statement

SK and MJZ report being employed by RainCity Analytics. JW, SWW, TI, JC, IK, and WP report being employed by Kite, a Gilead Company.

Data availability statement

All non-confidential data are available upon request.

Additional information

Funding

The study sponsor – Kite, a Gilead Company – provided funding for this study and was involved in the study design and preparation of the manuscript.