Tocilizumab prophylaxis for patients with multiple myeloma treated with bispecific antibodies Open Access

The emergence of US Food and Drug Administration (FDA)–approved bispecific antibodies for the treatment of relapsed and refractory multiple myeloma has significantly changed the treatment landscape.¹ Early studies of the FDA-approved B-cell maturation antigen (BCMA) × CD3– and G-protein–coupled receptor class C group 5 member D (GPRC5D) × CD3–targeted bispecific antibodies in relapsed and refractory multiple myeloma have shown high response rates at the expense of cytokine release syndrome (CRS), immune effector cell–associated neurotoxicity syndrome (ICANS), and infections.² Each of the available bispecific antibodies for multiple myeloma is associated with CRS/ICANS and comes with an FDA-mandated risk evaluation and mitigation strategy program to facilitate safe use of the drugs. The burden of CRS has been seen across the spectrum of hematologic malignancies, starting with chimeric antigen receptor (CAR) T-cell therapy. Based on the available limited data, in patients with hematologic malignancies who receive CAR T-cell therapy, CRS and ICANS are common causes of 30-day readmission, with a median of 5 additional hospital days and median charge of $64 561 per readmission.³ In patients with multiple myeloma treated with bispecific antibodies, we are now seeing the burden of increased number of hospital days, increased health care resource use and poor patient experience. Such associated costs include direct costs (ie, drugs), indirect costs (ie, loss of income), and management of treatment-related adverse events (ie, readmission for CRS).⁴ Novel approaches are needed to improve patient quality of life, resource use efficiency, and patient access to drugs.

The management of CRS commonly involves the use of steroids, antipyretics, interleukin-6 (IL-6) blocking agents (such as tocilizumab), and supportive/symptomatic care. These measures are commonly used in a reactive way once a patient is experiencing CRS. Substudies of larger trials have reported on the use of prophylactic tocilizumab before teclistamab (BCMA×CD3) and the investigational bispecific antibody cevostamab (Fc receptor-homolog 5 × CD3).^5,6 Limited real-world evidence is also emerging for prophylactic tocilizumab before teclistamab (BCMA×CD3).^7-9 To our knowledge, there is also a single-center study of prophylactic dexamethasone given on the day after each step-up dose in an attempt to mitigate CRS.¹⁰ To our knowledge there has not yet been any systematic study of this approach before elranatamab (BCMA×CD3), talquetamab (GPRC5D), or the investigational agent linvoseltamab (BCMA×CD3). There is also an ongoing prospective study (OPTec) evaluating the use of prophylactic tocilizumab before teclistamab (ClinicalTrials.gov identifier: NCT05972135).¹¹ We were motivated to administer a single prophylactic tocilizumab dose before the first step-up-dose only for our patients treated with any of the currently FDA-approved bispecific antibodies (teclistamab, elranatamab, and talquetamab) or the investigational agent linvoseltamab. The aims of the present study were to (1) investigate if there is any negative impact on response rates with the addition of a single dose of prophylactic tocilizumab, and (2) to determine the impact of prophylactic tocilizumab on the onset of CRS.

This single-center real-world study includes all patients treated with prophylactic tocilizumab before teclistamab, elranatamab, linvoseltamab (on an investigator-initiated trial), talquetamab, or a combination at the University of Miami Hospital and Clinics, Sylvester Comprehensive Cancer Center between the first FDA approval (25 October 2022) and data cutoff (16 December 2024) (Table 1). Only patients treated with one of these drugs on a clinical trial disallowing the use of prophylactic tocilizumab were excluded. All data were collected on an institutional review board–approved protocol. The prophylactic dose of tocilizumab was administered as 1 single, 8 mg/kg dose (Figure 1). This single dose of tocilizumab was given only before the very first step-up dose (after the FDA-recommended premedications) and was not repeated before any subsequent step-up doses or future doses (Figure 1). Patients were hospitalized in accordance with respective risk evaluation and mitigation strategy programs and recommendations in United States Prescribing Information (USPI) for FDA-approved bispecific antibodies. Patients were admitted for expedited ramp-up of 4 days for elranatamab and 7 days for teclistamab or talquetamab. At the time of admission, all patients received infectious prophylaxis per institutional standard (Table 2). Other supportive care was administered when indicated at the judgment of the treating physician and commonly included IV immunoglobulin and granulocyte colony-stimulating factor. The antifungal fluconazole was only rarely given for prolonged neutropenia. The primary aim of the study was to first determine whether there was any alteration of response with the addition of prophylactic tocilizumab and then to investigate the effect on CRS mitigation, which was assessed by the standard Lee American Society for Transplantation and Cellular Therapy criteria (Lee et al¹²). ICANS was also assessed by the American Society for Transplantation and Cellular Therapy criteria. Other adverse events were graded according to Common Terminology Criteria for Adverse Events, version 5.0. Patients who experienced CRS after prophylactic tocilizumab were managed in accordance with our institutional standard (Table 3).

At the time of data cutoff, 119 patients with multiple myeloma were given prophylactic tocilizumab before treatment with an FDA-approved bispecific antibody (teclistamab, elranatamab, and talquetamab), a combination of 2 bispecific antibodies, or on an institutional investigator-initiated trial of linvoseltamab (ClinicalTrials.gov identifier: NCT06376526). Of these patients, 45 were treated with teclistamab, 40 were treated with elranatamab, 10 were treated with linvoseltamab, 23 were treated with talquetamab and 1 was treated with the combination of teclistamab and talquetamab as was done in a recently published study (Table 1).¹³ The median age was 67 years (range, 35-87), and 29 patients (24.6%) were aged ≥75 years. Of all patients, 62 (52.1%) were female, 22 (18.5%) were Black, and 49 (41.2%) were of Hispanic ethnicity. Most patients had a good Eastern Cooperative Oncology Group performance status score of 0 to 1 (70.6%), but we included patients with a performance status of 2 (23.5%) and 3 to 4 (5.9%). Seventy-six patients (63.9%) met at least 1 exclusion criterion for their respective drug’s clinical trial (common reasons for exclusion included: renal dysfunction, age, previous therapies, hematologic parameters, and Eastern Cooperative Oncology Group performance status, among others). Overall, 32 patients (26.8%) had high-risk disease biology defined by presence of del17p, t(4;14), or t(14;16). A total of 29 patients (24.8%) had myeloma bone marrow infiltration of >60%, with 1 patient that did not have a bone marrow biopsy within 1 year who was excluded. Patients received a median of 4.5 (range, 1-16) previous lines of therapy. Most patients were triple-class (89.1%) and penta-drug exposed (53.8%).

Overall, we observed a low rate of CRS (10.1%; binomial 95% confidence interval [CI], 5.3-17). The rate of CRS for teclistamab, elranatamab, linvoseltamab, and talquetamab was 8.9%, 12.5%, 0%, and 13%, respectively (Figure 1). The single patient treated with the combination of teclistamab and talquetamab did not experience CRS. Among patients who experienced CRS, 10 of 12 events were grade 1, with the remaining 2 events grade 2. The overall rate of grade 2 CRS was 1.7%. The rate of grade 2 CRS for teclistamab, elranatamab, linvoseltamab, and talquetamab was 0%, 2.5%, 0%, and 4.3%, respectively. In patients who experienced CRS, most events occurred during step-up dosing (3 after dose 1, 5 after dose 2, and 2 after dose 3). Two events were recorded after cycle 1 day 1 but before cycle 1 day 8). There were no grade ≥3 CRS events. There were no episodes of recurrent CRS. No additional doses of tocilizumab or dexamethasone were given for CRS. There was a total of 2 delays in therapy because of CRS (1 day and 2 days, respectively).

The authors observed some differences in baseline characteristics when comparing patients who did not experience CRS with those who had an episode of CRS despite prophylactic tocilizumab, although these differences are limited by the small sample size of patients experiencing CRS. Patients who experienced CRS were more likely to be of Black race (41.7% vs 16%) or Hispanic ethnicity (33.3% vs 23.6%). They were also more likely to have a bone marrow plasma cell percentage of >60% (58.3% vs 21%), lower median number of previous therapies (4 vs 5), less likely to have had a stem cell transplant (33.3% vs 40.6%), or less likely to have been exposed to a BCMA antibody–drug conjugate (0% vs 8.5%). However, patients who experienced CRS despite prophylactic tocilizumab were more likely to be triple-class exposed (100% vs 87.7%). Penta-drug exposure was similar in both groups (50% vs 53.8%).

The observed rate of ICANS was also low overall (5.9%; 95% CI, 2.4-11.7). The rate of ICANS for teclistamab, elranatamab, linvoseltamab, and talquetamab was 6.7%, 0%, 0%, and 17.4%, respectively. The single patient treated with the combination of teclistamab and talquetamab did not experience ICANS. The overall rate of grade ≥2 ICANS was 4.2%. The rate of grade 2 ICANS for teclistamab, elranatamab, linvoseltamab, and talquetamab was 2.2%, 0%, 0%, and 17.4%, respectively. The overall rate of grade ≥3 ICANS was 2.5%. All 3 grade 3 ICANS events were in patients treated with talquetamab (13%). The rate of recurrence for ICANS was 0.8%. Two repeat doses of tocilizumab were given for ICANS, additional dexamethasone was given to 4 patients for ICANS and 2 patients received anakinra for ICANS. Among patients who experienced ICANS, 2 were maximum grade 1, 2 were grade 2, and 3 were grade 3. One of the suspected grade 3 ICANS events (decreased immune effector cell encephalopathy score with possible seizure) occurred in a patient with known central nervous system myeloma and intracranial hemorrhage. The second suspected grade 3 ICANS event (seizure) occurred in a patient with plasma cell leukemia who required intensive care unit care because of pneumonia, dialysis dependence, and suspected disseminated Varicella-Zoster virus. The third suspected grade 3 ICANS event (decreased immune effector cell encephalopathy score) occurred in a patient with baseline dialysis requirement and neuropathy who had >90% plasmacytosis on most recent bone marrow biopsy. The complexity of these cases illustrates the real-world experience; in each case, ICANS could not be excluded.

Both hematologic and infectious toxicity were commonly observed in our patients, although this is consistent with other real-world populations treated with bispecific antibodies without prophylactic tocilizumab. Grade ≥3 neutropenia occurred in 53 patients (44.5%) within a median of 16 days (range, 1-261). Grade ≥3 anemia (35.3%) and thrombocytopenia (26.9%) were also commonly observed. Serum immunoglobulin G of <400 mg/dL occurred in 95 patients (79.8%) and documented infections occurred in 56 patients (47.1%) during a median follow-up period of 191.5 days. Grade ≥3 infections occurred in 21 patients (17.6%). Commonly observed types of infection included upper respiratory (23 [19.3%]), COVID-19 (11 [9.2%]), pneumonia (8 [6.7%]), cytomegalovirus reactivation (7 [5.9%]), and neutropenic fever (6 [5%]).

Among the 105 response evaluable patients with a diagnosis of multiple myeloma, the overall response rate (ORR) was 69 of 105 (65.7%; 95% CI, 55.8-74.7). Patients were considered as not response evaluable for nonsecretory myeloma, or for those who received linvoseltamab as part of a separate clinical trial. The best objective response was complete response in 40 patients (38.1%; 95% CI, 28.8-48.1), partial response in 18 (17.1%; 95% CI, 10.5-25.7), stable disease in 22 (21%; 95% CI, 13.6-30), and progressive disease in 14 (13.3%; 95% CI, 7.5-21.4). A total of 51 patients (48.6%; 95% CI, 38.7-58.5) achieved a very good partial response or better. Median follow-up was 191.5 days (6-647) and among responders, the median duration of response was not reached. Patients with a short follow-up include those who died shortly after initiating therapy. At the time of data cutoff, 70 patients (58.8%) remain on bispecific antibody therapy. For individual drugs the ORR was 55.6%, 73%, and 78.3% for teclistamab, elranatamab, and talquetamab, respectively. The median progression-free survival was 8.1 months for teclistamab and not reached for elranatamab and talquetamab.

In this real-world study including 119 patients treated with BCMA×CD3- or GPRC5D×CD3-targeted bispecific antibodies for multiple myeloma, we found that a single dose of prophylactic IL-6–receptor inhibitor tocilizumab before the first step-up dose of bispecific antibodies translated into a very low rate of CRS (10.1%), mostly grade 1 events (and no grade 3), without the need for additional doses of tocilizumab or corticosteroids for CRS. Furthermore, this study provides evidence suggesting that a single dose of tocilizumab does not negatively affect the efficacy of bispecific antibodies. Our ORR (65.7%) is comparable with that of other real-world patient populations not treated with prophylactic tocilizumab.¹⁴ In addition, Figure 1C illustrates our observed ORRs, which do not differ substantially from those reported in the respective clinical trials for each drug in the absence of prophylactic tocilizumab. Our findings are in the context of the present real-world patient population, of whom 63.9% of our patients would have been excluded from their bispecific antibody’s respective trial for ≥1 comorbidities. Real-world patient populations are known to have higher rates of comorbid conditions than patients enrolled in typical clinical trials.

In clinical medicine there is often active discussion on the use of preventive vs reactive measures to manage drug toxicities. In this particular case, this real-world study demonstrates that tocilizumab prophylaxis reduces CRS seen with bispecific antibodies in multiple myeloma (Figure 1). Our real-world evidence results suggest that tocilizumab may be effective as a preventive, rather than reactive, measure. Furthermore, we observed an overall rate of CRS that is lower than registrational clinical trials for these agents (Figure 1).^15-18 The low CRS rate of 10.1% of patients represents a substantial improvement on the safety profile of this class of medications. Our findings are consistent with other smaller patient cohorts treated with prophylactic tocilizumab.^5,7-9 To our knowledge, this is the largest cohort of patients given prophylactic tocilizumab before antimyeloma bispecific antibodies to date and the first to include bispecific antibodies of multiple targets. Despite the fact that this is the largest study to date, in view of the differences in step-up dosing with the individual bispecific antibodies, pooled data from all bispecific antibodies should be interpreted with caution.

The impact of prophylactic tocilizumab on ICANS remains less clear. We observed a low rate of overall ICANS, however there were 3 cases of grade 3 ICANS. Patients treated with talquetamab had the highest observed rate of ICANS including all 3 grade 3 cases. It remains unclear whether this is because of the drug target itself (GPRC5D vs BCMA) or another reason. The overall impact of prophylactic tocilizumab on ICANS needs further investigation to determine whether this is a potential new safety signal or whether these events were related to the complexity of the clinical picture for each individual patient. It is also possible that previous studies have underreported ICANS because of clinically more significant and prevalent CRS symptoms, which may have masked coexisting onset of ICANS. Future studies are needed to better understand these aspects. Limitations to this real-world study include the lack of a placebo-controlled comparator arm.

The reduction in CRS in this study is an important observation because the onset of CRS has the potential to lead to increased hospital days, increased resource use, delays in treatment, and worse patient experience. This impact is of particular significance with grade ≥2 CRS, which we observed at a low rate (1.7%). When a patient is admitted for the management of CRS, and the medical team provides guideline concordant care, the admission will typically involve imaging, cultures, broad spectrum antibiotics, vital signs, and additional diagnostic workup on a case-by-case basis.² The cost of this workup is not negligible and requires added days in the hospital. In contrast, the improved safety profile seen in our study suggests that with prophylactic tocilizumab leads to decreased resource use and improved patient experience because fewer patients experienced CRS symptoms. On our study only 12 patients (10.1%) required additional workup for CRS, and none required additional doses of tocilizumab or dexamethasone for CRS. Despite most CRS events being grade 1 or 2 if tocilizumab is not given prophylactically, patients still often require a dose of tocilizumab to manage CRS. In the MajesTEC-1 trial of teclistamab, 60 patients (36.3%) required tocilizumab to manage CRS; and in 1 real-world study of teclistamab, 36% of patients required tocilizumab.^14,18 Similarly, tocilizumab was used to manage CRS in clinical trials with elranatamab (22.7%) and with talquetamab (58%).^16,17 In this study, we did not give any repeat doses of tocilizumab for CRS.

The current USPI for teclistamab recommends step-up dosing on days 1, 4, and 7, resulting in a 9-day total admission, with the condensed version allowing for a 7-day admission. Each of these timeframes assume no delay for toxicity, such as CRS. In a previous real-world study of 110 patients with multiple myeloma treated with teclistamab, authors observed a median admission duration of 9 days and a 29% readmission rate within 30 days of starting therapy, although this included admissions for other causes such as infection.¹⁴ In this real-world study including 119 patients with multiple myeloma we observed a shorter median hospitalization of with 7 days (range, 5-19) for teclistamab, mirroring the expedited step-up dosing schedule without delay. Similarly, our median time of admission mirrored the respective expedited step-up dosing in the USPI days for elranatamab (4 days [range, 3-46]) and for talquetamab (7 days [range, 6-39]). We also observed only 2 delays related to CRS (1 day and 2 days in duration) with no readmissions for CRS. The potential for improved efficiency of resource use through avoidance of treatment delays and reduced number of hospital days warrants further formal evaluation. Furthermore, in this study we did not give any repeat doses of dexamethasone for CRS. Prophylactic tocilizumab as a potentially steroid-sparing approach limits the burden of steroid-associated toxicities, infections, and interference with bispecific antibody efficacy. The administration of prophylactic tocilizumab has a substantial impact on the safety profile of bispecific antibodies and suggests an improvement in health care resource use.⁴ Furthermore, the identification in this study of differences in baseline characteristics, among patients not experiencing CRS and those experiencing CRS despite prophylactic tocilizumab, warrants further investigation and may be useful for patient selection in the future.

Although the USPI recommends hospitalization for step-up dosing for each of these drugs, the low rate and low severity of CRS in our study warrants further investigation of prophylactic tocilizumab as a tool to facilitate outpatient step-up dosing with no required hospitalization. In the authors’ opinion, this approach is safe for most patients in the setting of prophylactic tocilizumab. This is in line with other attempts to move step-up dosing completely outpatient with wearable monitoring technology, among others.¹⁹ As our community gains experience with these therapies, there will be patients with certain clinical pictures (eg, severe pain, underlying comorbidities, central nervous system myeloma, lack of caregiver support, and other complicating factors) who should still receive step-up dosing in the inpatient setting despite strategies to limit CRS. The outpatient step-up approach facilitated by prophylactic tocilizumab has the potential to improve patient experience and quality of life along with health care resource use efficiency. The present data also support the expansion of access to step-up dosing of these novel therapies into the community setting to increase patient access to effective novel therapies.

In summary, our real-world study shows that a single dose of prophylactic IL-6 receptor inhibitor tocilizumab before the first step-up dose of BCMA×CD3- or GPRC5D×CD3-targeted bispecific antibodies in multiple myeloma translated into a very low rate of CRS, mostly grade 1 events (and no grade 3), and without need for additional doses of tocilizumab or corticosteroids for CRS. Importantly, this approach also does not negatively affect the efficacy of bispecific antibodies. The improved safety profile decreases resource use, improves patient experience, and suggests that tocilizumab is effective as a preventive measure. The findings in this paper are important because they describe a prophylactic tocilizumab approach to substantially improve CRS rates and severity in a single disease state across therapies independent of drug target. The present cohort also includes a single patient treated with a combination of teclistamab and talquetamab at the same time who did not experience CRS after following our protocol. Our findings support the investigation of this approach in other hematologic malignancies and even in solid tumors in which T-cell engaging therapies are being used or under development in the drug discovery pipeline.

The authors thank the patients who participated in these studies and their families, and the staff members involved in data collection and analyses.

All authors are supported by Sylvester Comprehensive Cancer Center National Cancer Institute Core grant P30 CA 240139. In addition, O.L. is supported by the Riney Family Multiple Myeloma Research Program Fund, Tow Foundation, Myeloma Solutions Fund, and Cannon Guzy Family Fund.

Contribution: A.K. designed research, performed research, contributed data, analyzed data, wrote the draft of the manuscript, revised and finalized the draft, and approved the final version of the manuscript; J.L., B.D., D.C., M.K., F.M., J.H., A.P., and D.K., contributed data, analyzed data, gave scientific input, and approved the final version of the manuscript; O.L. designed research, performed research, contributed data, analyzed data, wrote the draft of the manuscript, revised and finalized the draft, and approved the final version of the manuscript; and all authors critically revised the manuscript, approved the final version for publication, and agreed to be accountable for all aspects of the work.

Conflict-of-interest disclosure: O.L. reports funding from National Cancer Institute/National Institutes of Health, US Food and Drug Administration, Leukemia and Lymphoma Society, Rising Tide Foundation, Multiple Myeloma Research Foundation, International Myeloma Foundation, Paula and Rodger Riney Foundation, Tow Foundation, Perelman Family Foundation, Myeloma Solutions Fund, Cannon Guzy Family Fund, Amgen, Celgene, Janssen, Takeda, Glenmark, Seattle Genetics, and Karyopharm; honoraria from AbbVie, Adaptive, Amgen, Binding Site, Bristol Myers Squibb, Celgene, Cellectis, GlaxoSmithKline (GSK), Janssen, Juno, and Pfizer; participated on advisory boards for AbbVie, Adaptive, Amgen, Binding Site, Bristol Myers Squibb, Celgene, Cellectis, GSK, Janssen, Juno, and Pfizer; and served on independent data monitoring committees for international randomized trials by Takeda, Merck, Janssen, and Novartis outside the submitted work. The remaining authors declare no competing financial interests.

Correspondence: Andrew Kowalski, Sylvester Myeloma Institute, Department of Pharmacy, Sylvester Comprehensive Cancer Center, University of Miami, 1475 NW 12th Ave, 2nd Floor, Miami, FL 33136; email: axk1682@miami.edu; and Ola Landgren, Sylvester Myeloma Institute, Division of Myeloma, Department of Medicine, Sylvester Comprehensive Cancer Center, University of Miami, 1120 NW 14th St, Clinical Research Building, Room 650D, Miami, FL 33136; email: col15@miami.edu.