Real-World Evidence Evaluating Teclistamab in Patients with Relapsed/Refractory Multiple Myeloma: A Systematic Literature Review

Simple Summary

Teclistamab belongs to a class of drugs known as bispecific antibodies. It is used to treat patients with relapsed or refractory multiple myeloma who have tried several other types of therapy. Teclistamab was initially studied in the MajesTEC-1 clinical trial and has been increasingly used in clinical practice since its approval in October 2022. In this study, we perform a systematic literature review to gather evidence on effectiveness, safety, healthcare resource utilization, and prescribing patterns associated with teclistamab in real-world observational studies. Results indicate that patients treated with teclistamab across health centers in the real-world have similar outcomes to patients treated in the clinical trial setting. As patients with relapsed or refractory multiple myeloma have a high disease burden and limited treatment options, results from this study will inform healthcare providers and health policy makers of the therapeutic benefits and safety profile of teclistamab in this patient population.

Abstract

Background: Teclistamab (TEC) is the first B-cell maturation antigen-directed bispecific antibody approved in 2022 by the European Medicines Agency and Food and Drug Administration for triple-class exposed relapsed/refractory multiple myeloma (RRMM). Objectives: As TEC is increasingly used in real-world (RW) settings, this study seeks to gather existing RW evidence on effectiveness, safety, healthcare resource utilization, and clinical practices associated with TEC. Methods: A systematic literature review was performed to identify RW observational studies of TEC-treated adults with RRMM from 2023 to June 2024. Results: Sixty-one records representing 41 unique studies were included; sample sizes ranged from 8 to 572 patients. Where reported, median follow-up ranged from 2.3 to 33.6 months, and >65% of the patients would have been ineligible for the pivotal trial of TEC (MajesTEC-1) in all but one study. In eight studies with ≥50 patients and ≥3 months follow-up, overall response rates were 59–66% and cytokine release syndrome (CRS) rates were 18–64%. Tocilizumab use for CRS management was reported in 14 studies, with two indicating CRS rates of 13% and 26% when used prophylactically. Survival and infection outcomes showed wide variability due to short follow-up in most studies. Conclusions: Overall, early RW effectiveness and safety outcomes of TEC were comparable to findings from MajesTEC-1.

1. Introduction

Multiple myeloma (MM), a plasma cell malignancy, is the second most common hematologic cancer in the United States (US), with an estimated 35,780 new cases (1.8% of all cancer diagnoses) and 12,540 deaths (2.0% of all cancer deaths) in 2024 [1]. While 5-year survival rates of MM have more than doubled since the 1970s to a rate of 61.1% [1,2], the disease is typically characterized by a series of responses and relapses meaning patients ultimately require multiple lines of therapy [3]. It imposes a substantial healthcare burden, as patients progress through successive lines of therapy; a recent retrospective database study found that all-cause healthcare costs averaged USD 35,760 per patient per month among patients with four or more prior lines of therapy [4]. After the three main classes of drugs for MM (proteasome inhibitors [PI], immunomodulatory imide drugs [IMiD], and anti-CD38 monoclonal antibodies [CD38 mAb]) have failed patients, few effective options remain for relapsed/refractory multiple myeloma (RRMM). Deciding on the next steps in treatment requires clinicians and patients to consider the pace of disease progression, patient comorbid conditions, potential clinically relevant toxicities, and the timing as well as financial implications of treatment [5].

The treatment landscape for RRMM has evolved considerably with the introduction of innovative targeted therapies, including B-cell maturation antigen (BCMA)-targeted chimeric antigen receptor (CAR) T-cell and bispecific therapies since 2020 [6]. Teclistamab-cqyv (TECVAYLI^®, Janssen Biotech, Horsham, PA, USA) is the first-in-class BCMA-targeted bispecific antibody (BsAb) approved by the European Medicines Agency (EMA) and the Food and Drug Administration (FDA) in August and October 2022, respectively, for adult patients with RRMM who have previously received at least three prior lines of therapy in Europe or at least four prior lines of therapy in the US, including a PI, IMiD, and CD38 mAb.

Teclistamab (TEC) is administered subcutaneously, initially using a step-up dosing (SUD) schedule to lower the risk of cytokine release syndrome (CRS) and neurotoxicity, followed by weight-based weekly or every-other-week treatment doses; patients are recommended to be hospitalized for 48 h after SUD for monitoring. Although TEC SUD is recommended to be delivered in an inpatient setting, since its launch, clinicians have been exploring different care models, including outpatient administrations, with the goal of reducing healthcare resource utilization (HCRU) and improving treatment experiences while ensuring patient safety [7,8,9].

MajesTEC-1, the initial phase I/II trial reporting on the efficacy and safety of TEC, demonstrated an overall response rate (ORR) of 63% [10,11], including 46% of patients achieving complete response (CR) or better [11]. CRS occurred in 72% of patients (most CRS events were grade 1 or 2 in severity while one patient [0.6%] experienced a grade 3 event in the setting of an ongoing infection), and 3% developed immune effector cell-associated neurotoxicity syndrome (ICANS) [10]. MajesTEC-1 utilized stringent eligibility criteria that excluded patients with significant disease burden, such as those with organ dysfunction, poor performance status, serious comorbid conditions, severe cytopenia, and/or prior BCMA-directed therapy exposure; however, these characteristics are common in real-world (RW) practice [10,12]. As a result, patients treated in MajesTEC-1 may not have been representative of all indicated patients in a RW setting [12]. Several consortia and institutions have disseminated reports on early RW outcomes and safety with TEC. As TEC is increasingly being used in RW settings with a growing amount of RW evidence being published, this systematic literature review (SLR) aims to identify and summarize the latest RW outcomes of TEC, including effectiveness, safety, healthcare provider practices, and associated HCRU.

2. Materials and Methods

This SLR was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [13]. This review has not been registered. The full study protocol and checklist are provided in Table A1 and the Supplementary Materials for reference.

Study eligibility criteria were defined in terms of the population, intervention, comparator, outcome, and study design (PICOS) structure outlined in Table A1 which guided the identification and selection of studies for the SLR. The target population included adult patients (≥18 years) with MM. The only intervention of interest was TEC, but no restrictions were applied to comparators. No restrictions were applied to outcomes to ensure that all relevant evidence was captured. Observational RW studies (prospective and retrospective) were included, while clinical trials, pooled analyses of clinical trials, case reports, case series, and narrative reviews were excluded. Publications of SLRs were excluded, but the bibliographies of relevant SLRs were screened for any relevant citations not otherwise captured in the primary searches (i.e., these references served only as secondary sources to ensure that all studies meeting the eligibility criteria were identified). Only English-language publications of full-text articles and conference materials (i.e., abstracts, posters, or oral presentations) were included, and a time restriction from 2023 to 2024 was applied.

Relevant studies were identified by searching the following databases through the Ovid platform: Medical Literature Analysis and Retrieval System Online (MEDLINE) and Excerpta Medica database (Embase). The specific search algorithms included a combination of indexing and free-text terms (see search terms in Table A2 and Table A3). The population terms were adapted from existing reviews [14,15], and terms for the generic and brand names of TEC were incorporated.

The main database searches were augmented with searches of 15 specific clinical or managed care conference proceedings from 2023 and 2024 (Table A5). The Northern Light database was used to search for studies from conferences that were indexed in the database (see search terms in Table A4), while conference websites were hand searched for the remaining conferences. Database searches were conducted at the end of May 2024, and further hand searches were conducted through the end of June 2024 to ensure all relevant materials were captured through the first half of 2024.

One reviewer performed all citation screening and data extraction, with quality checks performed by a second reviewer; a third reviewer was included to reach a consensus on any remaining discrepancies, where necessary. Screening decisions and extracted data were stored and managed in Microsoft Excel (Microsoft, Redmond, WA, USA). The study identification and selection process was summarized with a PRISMA flow diagram [13]. Kaplan–Meier (KM) curves for progression-free survival (PFS) and overall survival (OS) were extracted using DigitizeIt (DigitizeIt, Braunschweig, Germany) and reconstructed using an algorithm published by Guyot et al., 2012 [16]. The Newcastle–Ottawa Scale (Ottawa Hospital Research Institute, Ottawa, ON, Canada) was used to assess the quality of observational studies where full-text publications were available (Table A6). The scale utilizes a ‘star system’ to judge (i) the selection of the study groups, (ii) the comparability of the groups, and (iii) the ascertainment of either the exposure or outcome of interest for case–control or cohort studies, respectively.

3. Results

3.1. Study Characteristics

Of the 156 records identified across all sources, a total of 61 publications (five full-texts [12,17,18,19,20] and 56 conference abstracts, posters, or oral presentations) representing 41 unique studies evaluating TEC in a RW setting were identified (Figure 1). The included studies spanned seven countries in North America and Europe. Thirty-five of the 41 studies were retrospective in design. Fifteen studies (36.6%) reported results pooled from multiple centers, while 25 studies (61.0%) reported experience from a single center. Most studies (n = 34) were chart reviews, and seven reported data gathered from secondary databases. In 40 studies, sample sizes ranged from 8 to 572 patients. One study using the FDA Adverse Event Reporting System database reported 719 adverse event (AE) cases of TEC, rather than the event rates among patients [21], and, therefore, was not included in the results summary. Median follow-up (mFU) duration was reported in 19 studies (46.3%) and ranged from 2.3 [22] to 33.6 months [23]. This current review focused on eight studies with sample sizes ≥50 (range: 52 to 419) and mFU ≥3 (range: 3.1 to 9.5) months, as studies with larger sample sizes may be more generalizable to the wider population and those with longer follow-up duration may help to accurately evaluate AEs and outcomes (the median time to CR or better was 4.6 months [24] in MajesTEC-1). Exceptions were made for seven studies in order to report on data from novel topics including prophylactic tocilizumab (TCZ) use, less frequent dosing, and step-up dosing in the outpatient setting that have not been extensively evaluated or reported in the real-world yet and where data are primarily limited to smaller sample sizes with shorter follow-up periods. Studies not reporting mFU were not included in this synthesis due to the limited interpretability of the outcomes.

Figure 1. PRISMA flow diagram of included studies. Abbreviations: ASCO, American Society of Clinical Oncology; ASH, American Society for Hematology; EHA, European Hematology Association; HOPA, Hematology/Oncology Pharmacy Association; IMS, International Myeloma Society; LL&M, Lymphoma, Leukemia & Myeloma Congress; ONS, Oncology Nursing Society; SOHO, Society of Hematologic Oncology.

Study characteristics of studies with ≥50 patients and ≥3 months mFU are presented in Table 1 (Table A7 summarizes all 41 included studies). ORR and CRS rates were the most frequently reported effectiveness and safety measures, respectively. AE management using TCZ, length of hospital stay during TEC SUD, ICANS rates, and infections were also frequently reported. Common subpopulations included patients with prior anti-BCMA exposure and those receiving outpatient SUD. Studies that included other interventions alongside TEC were included when TEC-only subpopulation data were available (labeled as “TEC treatment group” in subsequent tables). Based on the Newcastle–Ottawa scale, the five included full-text publications [12,17,18,19,20] were high quality regarding study population selection and ascertainment of outcomes. However, all five studies had a non-comparative design and could not be compared directly.

Table 1. Summary of study characteristics and outcomes evaluated in studies with ≥50 patients and ≥3 months median follow-up.

Study ID	Study Design	Data Source	Chart Review vs. Secondary Database	Setting	Study Timeframe	mFU (Months)	Sample Size	Relevant Outcomes Evaluated
Banerjee 2023b [25,26]	Retrospective	Acentrus MM electronic medical records (EMRs)	Secondary database	Academic centers/community-based hospitals, multi-center	October 2022–November 2023	5.1	247	LOS, CRS, ICANS, step-up dosing schedule, time to next treatment (TTNT), or death
Dima 2023 [12,27,28,29,30]	Retrospective	Patients treated at USMIRC centers	Chart review	Academic, multi-center	August 2022–August 2023	3.8	106	CRS, ICANS, infections, ORR, DOR, OS, PFS, TCZ use in AE management, hospitalizations
Firestone 2023 [17,31,32,33]	Retrospective	Patients treated at the Memorial Sloan Kettering Cancer Center	Chart review	Academic, single-center	November 2022–July 2023	3.1	52	Survival, PFS, ORR, safety
Mohan 2024 [19,34]	Retrospective	Patients treated at five US academic centers	Chart review	Academic, multi-center	NR	3.5	110	CRS, ICANS, infections, ORR, best response, LOS, IVIG use in AE management, TCZ use in AE management
Pianko 2024 [35]	Retrospective	Komodo Healthcare MapTM	Secondary database	Multi-center	October 2022–December 2023	4.2	419	Dosing schedule of teclistamab, time to less frequent dosing, TTNT
Riedhammer 2024 [20,36]	Retrospective	Patients treated at 18 German centers	Chart review	Multi-center	July 2022–October 2023	5.5	123	Time to response, best response, ORR, PFS, infections
Tan-Asoori 2023 [37,38,39]	Retrospective	Patients treated at IMF-associated centers	Chart review	Academic, multi-center	NR—October 2023	5	204	ORR, OS, PFS, CRS, ICANS, TCZ use in AE management, IVIG use in AE management, LOS
Tan 2024 [40,41]	Retrospective	Patients treated at the Memorial Sloan Kettering Cancer Center	Chart review	Academic, single-center	November 2022–March 2024	Overall population: 9.5 Patients switching to less-frequent dosing: 6.4	86	ORR, DOR, PFS, time to response

Abbreviations: AE, adverse event; CRS, cytokine release syndrome; DOR, duration of response; EMR, electronic medical record; ICANS, immune effector cell-associated neurotoxicity syndrome; IMF, International Myeloma Foundation; IVIG, intravenous immunoglobulin; LOS, length of stay; mFU, median follow-up; MM, multiple myeloma; NR, not recorded; PFS, progression-free survival; ORR, overall response rate; OS, overall survival; TCZ, tocilizumab; TTNT, time to next treatment; US, United States; USMIRC, US Myeloma Innovations Research Collaborative.

3.2. Population Characteristics

Population characteristics of studies with ≥50 patients and ≥3 months mFU are reported in Table 2 and Table 3. Summaries of all included studies are available in Table A8 and Table A9, with additional characteristics provided in Table A12 and Table A13. Within the overall populations, the median age ranged from 65 [35] to 71 [40] years, and 63.4% [35] to 76.0% [40] of the patients were White. High-risk cytogenetic abnormalities were reported in six studies and were present in 33% [17] to 71% [40] of the patients. Five studies consistently defined high-risk cytogenetics as having one of the following abnormalities: del17p, t(4;14), or t(14;16) [12,19,20,37,40]; four of these studies included 1q gain/amp [12,19,37,40]; two studies included t(14;20) [37,40]; and one study included monosomy 17 [37]. The remaining study did not provide a definition [17].

Table 2. Summary of patient characteristics in studies with ≥50 patients and ≥3 months median follow-up.

Study ID	Overall/Subgroup Details	Sample Size	Median Age (Range), Years	Female, n (%)	Race, n (%)			High - Risk Cytogenetics, n (%)	Median Prior Lines of Therapy	Extramedullary Disease, n (%)
					White	Black	Asian
Banerjee 2023b [25,26]	Overall	247	69 (41–89)	111 (44.9)	138 (75.8) a	23 (12.6) a	21 (11.5) a	--	--	14 (5.7)
Dima 2023 [12,27,28,29,30]	Overall	106	66.5 (35–87)	55 (54) b	72 (68)	28 (26)	2 (2)	56 (59)	6	45 (42)
Firestone 2023 [17,31,32,33]	Overall	52	70 (30–80)	--	--	--	--	17 (33)	7	18 (35)
Mohan 2024 [19,34]	Overall	110	68 (37–89)	54 (49)	67 (61)	32 (29)	2 (1.8)	59 c (62)	6	48 (44)
Pianko 2024 [35]	Overall	419	65 (58–73) d	183 (43.7)	185 (63.4) e	91 (31.2) e	16 (5.4) e	--	5	--
Riedhammer 2024 [20,36]	Overall	123	67 (35–87)	53 f (43.1)	--	--	--	39 (36.8) g	6	43 (36.1)
Tan-Asoori 2023 [37,38,39]	Overall	204	66 (33–91)	91 (45)	143 (70)	15 (7)	19 (9)	90 (44)	6	38 (19)
Tan 2024 [40,41]	Overall	86	71 (64–78) d	44 (51)	65 (76)	14 (16)	--	56 (71) h	6	30 (38) h

Notes: Double dashes (“—”) indicate data not reported. ^a Evaluated population = 186; ^b calculated from the reported proportion of males; ^c n = 95 evaluable; ^d IQR; ^e evaluable population, n = 292; ^f n hand calculated; ^g evaluable population, n = 106; ^h evaluable population, n = 79. Abbreviations: EMD, extramedullary disease; IQR, interquartile range.

Table 3. Summary of MajesTEC-1 ineligibility in studies with ≥50 patients and ≥3 months median follow-up.

Study ID	Overall/Subgroup Details	Sample Size	MajesTEC-1 Ineligible Patients, n (%)	Prior BCMA Therapy, n (%)	ECOG PS ≥2, n (%)	Cytopenia, n (%)				Renal Impairment/Failure, n (%)	CrCl <30 mL/min or 40 mL/min, n (%)
						Overall	Anemia	Neutropenia	Thrombocytopenia
Banerjee 2023b [25,26]	Overall	247	--	48 (19.4)	--	--	126 (51.0)	55 (22.3)	--	100 (40.5)	--
Dima 2023 [12,27,28,29,30]	Overall	106	88 (83)	56 (53)	35 (33)	33 (31)	27 (25)	2 (2)	21 (20)	14 (13)	14 (13)
Firestone 2023 [17,31,32,33]	Overall	52	--	27 (52)	--	--	--	--	--	--	--
Mohan 2024 [19,34]	Overall	110	--	38 (35)	--	--	--	--	--	--	--
Pianko 2024 [35]	Overall	419	--	102 (24.3)	--	--	164 (39.1)	50 (11.9)	--	206 (49.2)	--
Riedhammer 2024 [20,36]	Overall	123	48 a (39)	45 (37.4)	--	--	--	--	--	--
Tan-Asoori 2023 [37,38,39]	Overall	204	122 (70) b	91 (45.0)	--	--	--	--	--	--	25 (12)
Tan 2024 [40,41]	Overall	86	--	32 (37)	5 (10) c	--	--	--	--	--	9 (10)

Notes: Double dashes (“—”) indicate data not reported. ^a n hand calculated; ^b based on the evaluated population (n = 175); ^c based on the evaluable population, n = 50. Abbreviations: BCMA, B-cell maturation antigen; CrCl, creatinine clearance; ECOG, Eastern Cooperative Oncology Group performance status; EMD, extramedullary disease.

Extramedullary disease (EMD) was present in 19% [37] to 44% [19] of the patients in six chart review studies, where documented. One study utilizing data from a multi-center MM electronic medical record (EMR) database reported that EMD diagnosis codes were present in 5.7% of the patients [25]. Of these studies, two defined EMD as any plasmacytomas [17,25] and four defined EMD as plasmacytomas not associated with the bone [12,19,37,40]. Only one study did not provide a definition for EMD [20]. The median number of prior lines of therapy ranged from five [35] to seven [17], such therapies including patients with prior BCMA-directed therapy exposure. Other prior lines of therapy for all included studies can be found in Table A12.

MajesTEC-1 ineligibility was reported in three studies: ≥70% of patients were ineligible in two studies [12,37] and one study reported 39% of the patients as ineligible [20]. The most common reasons for ineligibility in these studies included prior BCMA-directed therapy (37% [20] to 53% [12]), a poor performance status (Eastern Cooperative Oncology Group [ECOG] ≥2; 33% [12]), cytopenia(s) (31% [12]), and/or renal impairment/failure (12% [37] to 13% [12]). Additionally, individual cytopenia ranges across studies with ≥50 patients and ≥3 months mFU included anemia (25% [12] to 51% [25]), neutropenia (2% [12] to 22% [25]), and thrombocytopenia (20% [12]).

3.3. Outcomes

The following sections summarize the most frequently reported outcomes of interest for the overall populations in the included studies with ≥50 patients and ≥3 months mFU, focusing on the latest timepoint in each study. Table A10 and Table A11 provide outcome summaries for all included studies.

3.3.1. Effectiveness Outcomes

Six of the eight studies with ≥50 patients and ≥3 months mFU reported on effectiveness outcomes (Table 4), and variability in response and survival rates was observed across these studies. Among these studies, ORR (partial response [PR] or better) ranged from 59% [20] to 66% [12] (n = 6 studies), very good partial response (VGPR) or better ranged from 38% [17] to 51% [19] (n = 6 studies), and CR or better ranged from 19% [37] to 29% [12] (n = 4 studies).

Table 4. Summary of key effectiveness outcomes in studies with ≥50 patients and ≥3 months median follow-up.

Study ID	Overall/Subgroup Details	Timepoint/mFU	Response Evaluable Population Size	Overall Response Rate, n (%)			Survival Evaluable Population Size	PFS		OS
				PR or Better	VGPR or Better	CR or Better		Median, Months (95% CI)	6-Month Rate, % (95% CI)	Median, Months (95% CI)	6-Month Rate, % (95% CI)
Dima 2023 [12,27,28,29,30]	Overall	Median of 3.8 months	104	70 a (66)	49 (46) b	31 (29) b	106	5.4 (3.4, NR)	--	NR	70 (61, 80)
	>70 years old	Median of 3.8 months c	34	24 (71)	--	10 (30) d	33	5.4 (2.8, NR)	--	--	--
Firestone 2023 [17,31,32,33]	Overall	Median of 3.1 months	47	30 (64)	18 a (38)	--	52	NR	--	--	--
	Prior anti-BCMA exposure	Median of 3.1 months c	26	13 (50.0)	--	--	27	3.4	--	--	--
Mohan 2024 [19,34]	Overall	Median of 3.5 months	98	61 (62)	50 a (51)	20 a (20)	110	NR	52 (42, 64)	NR	80 (72, 89)
Riedhammer 2024 [20,36]	Overall	Median of 5.5 months	123	73 a (59.3)	59 a (48.0) b	27 a (22.0) b,e	123	8.7	--	NR	--
Tan-Asoori 2023 [37,38,39]	Overall	Median of 5 months	180	115 a (64)	90 a (50) b	34 a (19) b	204	13	57.7	15	76.2
Tan 2024 [40,41]	Overall	Median of 9.5 months	77	47 (61)	33 (43)	--	86	--	52.4 (42.4, 64.7)	--	--
	Prior BCMA-directed therapy	Median of 6.4 months	32	14 (43)	--	--	32	--	90.0 (79.1, 100.0)	--	--

Notes: Double dashes (“—”) indicate data not reported. ^a n hand calculated; ^b calculated from stratified best response; ^c median follow-up assumed to apply to subgroup data, as both share the same data cut-off; ^d % hand calculated; ^e reported as near complete or complete response. Abbreviations: BCMA, B-cell maturation antigen; CI, confidence interval; CR, complete response; mFU, median follow-up; NR, not reached; PR, partial response; VGPR, very good partial response.

Where reported, PFS and OS varied across studies with ≥50 patients and ≥3 months mFU, as shown in Table 4. Among the studies reporting the median PFS (n = 5 studies; mFU 3.1 [17] to 5.5 [20] months), the median PFS ranged from 5.4 [12] to 13 [37] months and was not reached in two studies [17,19]. The 6-month PFS rate was similar across studies, where reported (52% [19] to 58% [37] in three studies; mFU 3.5 [19] to 9.5 [40] months). Among the studies reporting OS, the median OS was not reached in three studies (mFU 3.5 [19] to 5.5 [20] months), while one reported a median of 15 months (mFU 5 months [37]). The 6-month OS rate ranged from 70% [12] to 80% [19] (n = 3 studies; mFU 3.5 [19] to 5 [37] months).

3.3.2. Safety Outcomes

Safety outcomes were reported in six of the eight studies (Table 5), with variability in the proportion of patients experiencing AEs observed across all studies. All six studies reported CRS rates, while three studies reported ICANS rates and four studies reported infection rates in the overall population. Among these studies, the proportion of patients who developed any grade of CRS ranged from 18% [25] to 64% [12] (n = 6 studies), and a small proportion of patients (0.5% [37] to 4.5% [19]) experienced grade ≥3 CRS (n = 5 studies). When stratified by data source and care models, any grade CRS rates from chart reviews of patients with inpatient monitoring ranged from 52% [17] to 64% [12], one study reported a CRS rate of 36% [37] from chart reviews of patients with outpatient monitoring, and one study reported a CRS rate of 18.4% [25] from secondary databases (e.g., payer claims, EMRs), as identified by the International Classification of Diseases 10th Revision codes. Three studies reported any grade ICANS rates ranging from 4% [25] to 14% [12], and three studies reported grade ≥3 ICANS rates ranging from 0% [25] to 4.5% [19]. Any grade infections were experienced by 31% [12] to 60% [37] of the patients (n = 4 studies), and approximately 26% of the patients had grade ≥3 infections (n = 2 studies) [19,20]. Two studies reported the impact on CRS rates when using prophylactic TCZ [42,43], as described in more detail below (Section 3.3.3).

Table 5. Summary of key safety outcomes in studies with ≥50 patients and ≥3 months median follow-up.

Study ID	Overall/Subgroup Details	Sample Size	Timepoint/mFU	CRS, n (%)				ICANS, n (%)				Infections, n (%)
				Any Grade	Grade 1	Grade 2	Grade 3+	Any Grade	Grade 1	Grade 2	Grade 3+	Any Grade	Grade 3+
Chart review—mixed inpatients and outpatient, or not reported
Mohan 2024 [19,34]	Overall	110	Median of 3.5 months	62 a (56)	57 a (51.8) b,c		5 (4.5) b	12 (11)	--	--	5 (4.5)	44 (40)	29 (26)
Riedhammer 2024 [20,36]	Overall	123	Median of 5.5 months	72 (58.5)	--	--	2 (1.6)	--	--	--	--	67 (54.5)	33 (26.8)
Tan-Asoori 2023 [37,38,39]	Overall	204	Median of 5 months	110 (53.9) a,b	84 (41.2) a,b	25 (12.3) a,b	1 (0.5) a,b	--	--	--	--	115 (60.0)	--
Chart Review—inpatient monitoring
Dima 2023 [12,27,28,29,30]	Overall	106	Median of 3.8 months	68 (64)	57 (54)	10 (9)	1 (1)	15 (14)	5 (5)	7 (6)	3 (3)	33 (31)	--
	>70 years old	33	July 2023 cut-off	22 (67)	--	--	1 (3)	7 (21)	--	--	0 (0)	11 (33)	--
Firestone 2023 [17,31,32,33]	Overall	52	Median of 3.1 months	27 (52)	--	--	--	--	--	--	--	--	--
Tan-Asoori 2023 [37,38,39]	Inpatients	160	Median of 5 months	94 (59)	72 (45)	22 (14)	0 (0)	--	--	--	--	--	--
Chart review—outpatient monitoring
Tan-Asoori 2023 [37,38,39]	Outpatients	44	Median of 5 months	16 (36)	12 (27)	3 (7)	1 (2)	--	--	--	--	--	--
Secondary databases
Banerjee 2023b [25,26]	Overall	76	Median of 5.1 months	14 (18.4)	8 (10.5)	3 (3.9)	1 (1.3)	3 (3.9)	2 (2.6)	1 (1.3)	0 (0)	--	--

Notes: Double dashes (“—”) indicate data not reported. Results are based on the most recent timepoint in each study. ^a n hand calculated; ^b % hand calculated; ^c reported as grade 1 and 2. Abbreviations: CRS, cytokine release syndrome; ICANS, immune effector cell-associated neurotoxicity syndrome; mFU, median follow-up.

3.3.3. Healthcare Provider Practices and Resource Utilization

The most frequently reported outcome related to healthcare provider practices was TCZ use in CRS management, which was reported in 12 studies. Among the five studies with ≥50 patients and ≥3 months mFU, TCZ usage ranged from 15% [25] to 41% [12] in the overall population. Additionally, two studies with any sample size and any mFU reported on prophylactic TCZ use in the inpatient setting. Kowalski (2023) only included patients (n = 31) treated with prophylactic TCZ before the TEC treatment, and there was a CRS rate of 13% (all grade 1; 95% CI 4, 30; p < 0.01) [42]. Marin (2023) assessed patients treated with prophylactic TCZ prior to the second TEC SUD (n = 38) and compared them to a cohort of patients who received TEC without the prophylactic TCZ (n = 15) [43]. The cohort with the prophylactic TCZ had a 26% rate of any grade CRS, while the cohort without the prophylactic TCZ had a much higher rate of any grade CRS, at 73% [43]. Among the patients in the prophylactic TCZ cohort who experienced CRS events (n = 10), the majority of the events were grade 1 (n = 8), while only two patients had grade 2 or 3 events; CRS grades for the cohort without the prophylactic TCZ were not reported [43].

The most frequently reported HCRU outcome was hospital length of stay (LOS) during TEC SUD for inpatient TEC administration or AE management if hospitalized after an outpatient SUD administration. Three studies with ≥50 patients and ≥3 months mFU reported an inpatient SUD administration model, with the median LOS ranging from 8 days [37] to 9 days [12,34]. Two studies with any sample size and any mFU reported LOS data for 1-3-5 and 1-4-7 SUD schedules. Kawasaki (2024) reported a mean LOS of 7.6 days for the 1-3-5 schedule and 9.2 days for the 1-4-7 schedule. Graf (2024) reported a median LOS of 6 days for the 1-3-5 schedule and 9 days for the 1-4-7 schedule. Two studies reported that LOS for inpatient SUD decreased over time: Banerjee (2023a), using nationally representative payer claims data, reported a mean LOS of 11.4 (SD 9.0) days for patients who initiated TEC within the first four months of FDA approval (through February 2023), and a mean LOS of 7.0 (SD 1.4) days for patients who initiated TEC in the most recent month (July 2023 by the data cut-off) [44]. Similarly, Banerjee (2023b), using a multi-center MM EMR database, reported a median LOS of 11 days in December 2022 and a median LOS of 6.5 days in May 2023 [25]. Additionally, three studies (any sample size and any mFU) with patients undergoing outpatient SUD reported LOS for hospital admissions at any time due to AEs, with medians of 1.5 days [45], 2 days [46], and 4 days [47] per admission, respectively.

Three recently published studies with ≥50 patients and ≥3 months mFU reported the proportion and timing of patients switching from weekly dosing to less frequent dosing. Tan (2024) reported that 30 patients (34.9%) at a single academic center switched from every week (QW) to every 2 weeks dosing (Q2W) and two patients (2.3%) switched from QW to every 4 weeks (Q4W), with a median time to switch of 3.3 months, for the primary reason of achieving ≥PR (n = 23) and/or safety (n = 14) [40]. Among these patients who switched the dosing frequency, the 6-month PFS rate post-switch was 90% after an mFU of 6.4 months after switching [40]. Two studies, using nationally representative secondary data, estimated the probability and timing of patients switching to less frequent dosing using KM analysis. The probability of switching to less frequent dosing (32 out of 39 switchers went from QW to Q2W [26]; 58 out of 78 switchers went from QW to Q2W [35]) after three months was 15.5% [26] and 19.0% [35], and this increased to 38.3% [26] and 38.6% [35], respectively, at six months. The median time to switch was 8.5 months in one study [26] and it was not reached at an mFU of 4.2 months in the other [35].

4. Discussion

This is the first SLR to consolidate RW studies of TEC, globally. Wide variance was observed in the effectiveness and safety of TEC across the entire evidence base, with ORR ranging from 44% to 87%, CRS rates ranging from 6% to 85% depending on data sources and care models, and ICANS being present in 0% to 23% of the patients. When restricting the review to larger studies that included ≥50 patients with an mFU of ≥3 months, the range of results narrowed: ORR ranged from 59% to 66%, CRS rates ranged from 18% to 64% depending on data sources and care models, and ICANS rates ranged from 4% to 14%. Wide variability in PFS and OS estimates and limited data on infection were observed given the short follow-up time in most studies, and this trend was still observed when only considering studies with larger sample sizes. Early RW use of TEC was characterized by administration to a diverse group of patients, including minorities, patients with significant comorbid conditions, high-risk features, and those with prior BCMA-directed therapy exposure.

The collation of these data provides early insights into how RW outcomes of TEC compare with the results of the MajesTEC-1 trial (Table A15), taking into account that most RW studies report that more than half of the patients would have been ineligible for the MajesTEC-1 trial. Patients in the RW differ from the trial populations, since they are not required to meet stringent eligibility criteria and may have more comorbidities and less hematopoietic or organ reserve [10,12]. Additionally, with TEC being a first-in-class BCMA BsAb and with these studies capturing data from the first year since TEC approval, the patients included in RW studies may have had advanced diseases and may have been more heavily pretreated. Despite this, the early effectiveness and safety profile of TEC in RW practice appeared comparable with MajesTEC-1.

The ORR in MajesTEC-1 was 63% [10]; studies evaluating ORR in this SLR also provided similar results, with RW evidence showing a range of 59% [20] to 66% [12] among the studies with ≥50 patients and ≥3 months mFU. Despite the similarities in ORR, the depth of response differed in the early RW patients. In MajesTEC-1, 46% of the patients achieved a response of CR or better [11], whereas the CR rates ranged from 19% [37] to 29% [12] in the early RW setting. In RW practice, this may be a reflection of the hard-to-treat population which included heavily pretreated patients, high-risk features, significant comorbidities, and/or patients with prior exposure to BCMA-directed therapy, as well as the inability of the short follow-up period of the studies evaluated to assess CR or better. In RW practice, it may not be possible to undergo a bone marrow biopsy as frequently to confirm International Myeloma Working Group CR status, a fact which may result in the misclassification of responses as VGPR and result in an underestimated rate of CR or better. Furthermore, the median time to CR or better in MajesTEC-1 was 4.6 months [24], but, in RW studies, the follow-up was short, with only eight studies having an mFU ≥3 months, with a range of 3.1 to 9.5 months. This short follow-up in the RW studies may have limited the ability to assess the deepening of responses due to the delayed clearance of paraprotein following tumor killing [48], and thus may be unable to measure the best response, PFS, or OS effectively. Moreover, clarifying the duration between the patient’s prior BCMA-directed therapy and the first TEC dose in patients with prior exposure to BCMA-directed therapy will also be paramount to interpret their survival outcomes.

The rate of any grade CRS in the MajesTEC-1 clinical trial was 72% [10], which was greater than the range reported in RW studies with ≥50 patients and ≥3 months mFU (18% [25] to 64% [12]). However, the CRS rates in RW studies that used prophylactic TCZ (13% and 26%) [42,43] were similar to the subgroup in the MajesTEC-1 cohort that received prophylactic TCZ (26%) [24]. There was notable variation in CRS rates across the included studies, possibly due to differences in data sources, data collection methods, and the care models for CRS prophylaxis and monitoring. Patients found within secondary databases (e.g., payer claims, EMR) may have lower rates due to their reliance on diagnostic codes; for example, healthcare professionals may not code fever as CRS or code fever at all, and diagnosis codes specifically for CRS may not always be used. Chart reviews of patients with inpatient monitoring may represent a more rigorous definition of RW CRS rates with TEC, i.e., data are abstracted directly from physician’s notes, and CRS-relevant signs and symptoms are continuously monitored and recorded systematically in the hospital setting. However, it is important to note that these definitions may differ between settings and may not always be used. Data on patients who received outpatient monitoring tend to have lower rates compared to inpatient monitoring, as they may have different triggers and thresholds of CRS reporting and, sometimes, may rely on patient-reported symptoms. These differences in data sources and settings differ from the MajesTEC-1 trial as well, and, therefore, should be considered while interpreting results.

Studies included in the SLR also did not report if general neurotoxicity and ICANS were assessed together or if the reported value was for ICANS alone. Including neurotoxicity with ICANS may explain the higher rate of ICANS in the RW setting (4% [25] to 14% [12] among the studies with ≥50 patients and ≥3 months mFU) versus that observed in MajesTEC-1 (3%), as neurotoxicity symptoms cover a wide spectrum and definitions vary across practices.

Infection rates were sparsely reported across the RW studies, limiting the ability to analyze these data; among the larger studies with ≥3 months mFU, only four studies reported any grade infections (31% [12] to 60% [37]) and only two studies reported grade ≥3 infections (26% [19] and 27% [20]). MajesTEC-1 reported a higher rate of infections (79%), with 55% of the patients experiencing grade 3 or 4 infections [11], though this was in the context of longer follow-ups. Unlike CRS and ICANS, both of which occur at the beginning of the treatment and are less dependent on follow-up times in RW studies, infection risk persists throughout the treatment; therefore, studies with shorter follow-up times may underreport infection rates. Additionally, the enrollment of patients for MajesTEC-1 occurred concurrently with the onset of the COVID-19 pandemic, before the widespread vaccine availability and use, a fact which may have resulted in higher infection rates among MajesTEC-1 patients compared to the RW infection rates reported in the studies considered here, all of which were conducted in the post-pandemic period. Guidelines agreed upon by expert consensus have since been published, allowing for optimal treatment of infections during TEC treatment [49,50]. With the adoption of infection management guideline recommendations for MM, other reasons for lower RW infection rates could be routine vaccination against COVID-19 and prophylactic intravenous immunoglobulin use for hypogammaglobulinemia and antibiotic use. Given the short follow-up durations, these results must be interpreted with caution, and data on long-term infection prophylaxis and management will need to be assessed in future work.

This review also provides initial insights into healthcare provider practices and HCRU associated with TEC in the RW setting. The use of TCZ during SUD and other medications often used for managing AEs associated with TEC were frequently reported in the studies included in the SLR. Although preliminary results showed the benefits of using prophylactic TCZ in Marin (2023) [43] and Kowalski (2023) [42], further studies with larger sample sizes are warranted. Based on studies published in 2023 and the first half of 2024, most patients received TEC SUD in an inpatient setting, with some institutions starting to implement outpatient SUD models [7,9,37,45,51]. This is reflected in a recent study that conducted a panel interview of clinicians from 20 practices across 13 states who were among the first ones to start TEC in their practices after FDA approval. Within seven months of approval, 74% of these practices provided SUD exclusively in an inpatient setting, while 26% provided SUD in an outpatient or hybrid setting; all participating practices using inpatient SUD expressed desire of moving to outpatient SUD for BsAb in the future [52]. In two secondary database analyses that monitored inpatient LOS for SUD over time, a decreasing trend was observed [25,44]. This observation might be due to improved familiarity with TEC among providers over time, availability of well-established AE management protocols, and quality improvement models that institutions might have implemented to reduce HCRU. Furthermore, two studies using large databases of medical records suggest that patients were able to complete SUD without delay, with the two-day and three-day dosing schedules being the most common, and that the majority of the patients were taking pre-medications as recommended [25,53].

Recent studies with RW data available for patients who switched to a less frequent dosing schedule (most frequently switching to Q2W) have shown results in line with the MajesTEC-1 trial. In MajesTEC-1, patients were allowed to switch from weekly TEC doses to every-other-week if they achieved a PR or better after four or more cycles in phase 1 or a CR or better for six or more months in phase 2 [10,11]. With over two years of follow-up, TEC demonstrated deep and durable responses and reduced new onset grade ≥3 infections over time, aligning with the median time to switch to Q2W dosing (~11 month mFU), including in patients who switched to less frequent dosing [11]. The reduction in new onset grade ≥3 infections may also be impacted by the increasing usage of IVIG and prophylaxis [11]. Three RW studies identified in the SLR reported on less frequent TEC dosing, mainly with a Q2W schedule, among patients who switched [26,35,40]; Q4W was rarely observed within these studies. Tan (2024) reported a 6-month PFS rate of 90% in patients who switched to less frequent dosing based on response or for safety management [40]; longer follow-ups are needed to understand the long-term impact on PFS and OS.

This SLR has some limitations. First, the recent approval of TEC resulted in the restriction of studies to the 2023 and 2024 calendar years, leading to an evidence base consisting of mostly conference materials with only five full texts. As conference abstracts may not include as detailed information as full-text publications, these results should be interpreted with caution. Due to the publication date restriction, most studies included patients who initiated TEC within 1 year of approval. This resulted in short follow-up periods and small sample sizes; longer follow-up time is needed for RW outcomes to mature. Many of these early patients were expected to be sicker and with a higher disease burden, contributing to the high percentage of reported MajesTEC-1 ineligibility. Despite this, the early effectiveness and safety outcomes remain consistent with the clinical trial. When sample sizes allow it, results may be stratified by MajesTEC-1 eligibility to provide a more balanced comparison with clinical trial results. Further, given the short follow-up periods, outcomes such as PFS, OS, and infection rate often need to be considered carefully, as current data may not reflect true estimates over time. Finally, numbers at risk over time or the total number of events are needed for reasonable accuracy during KM curve extraction and recreation [16], but this information was not available in several studies. As a result, comparisons made through the KM curves are limited and need to be interpreted cautiously. These limitations are further affected by a wide variation in the data due to differences in sample sizes, differences in data sources, and differences in RW clinical practices across health centers. Due to these limitations, a meta-analysis with patient-level data would be an appropriate next step once the data mature.

5. Conclusions

In conclusion, this SLR summarizes the initial RW evidence available for TEC. Variation and immature data limit the interpretation of long-term outcomes; however, results from studies with larger sample sizes and ≥3 months mFU suggest that the early effectiveness and safety profiles for TEC in the RW are comparable to those from the pivotal trial, even for patients who were sicker, with a high disease burden, and which would not have met the eligibility criteria for MajesTEC-1.