Аномальные модели иммунной фиксации иммунной сыворотки широко распространены после BCMA CAR-T в множестве миеломы и предшествуют иммунным явлениям.

TO THE EDITOR:

Multiple myeloma (MM) is characterized by clonal plasma cell expansion associated with a monoclonal immunoglobulin that remains of the same isotype, with the exception of rare cases in which isotype switching occurs at progression of disease (POD). 1 , 2 However, when monoclonal immunoglobulins of a different isotype are identified, they are termed abnormal serum immunofixation patterns (ASIP). Studies have shown a range of how often ASIP occur in patients with MM depending on treatment, with as few as 11.8% or as many as 50.5% with ASIP. 3 , 4 , 5 , 6 , 7 Importantly, the presence of ASIP after autologous stem cell transplantation (ASCT) has been associated with longer overall survival, progression-free survival (PFS) and higher rates of complete response, 3 creating a rationale for examining this association after other therapies. ASIP after B-cell maturation antigen (BCMA) chimeric antigen receptor T-cell (CAR-T) therapy in relapsed/refractory MM has not been well characterized. Its success with overall response rates >70% and durability that exceeds those of many other therapies 8 , 9 , 10 , as well as its limitations with severe hypogammaglobulinemia being commonplace, makes it an ideal therapy to study possible prognostic markers. 11 , 12 To our knowledge, we present the largest cohort to date investigating ASIP, its association with survival, and the factors that may be related to ASIP occurrence, such as hypogammaglobulinemia and immune stimuli like infections and vaccinations.

All patients with relapsed/refractory MM who received ciltacabtagene autoleucel or idecabtagene vicleucel at 2 institutions (Weill Cornell Medicine or Columbia University Irving Medical Center) were included through June 2024. This study was conducted with institutional review board approval from both institutions in accordance with the Declaration of Helsinki. Details on methodology are included in supplemental Methods .

In total, 58 patients were included, with 15 patients receiving idecabtagene vicleucel (26%) and 43 receiving ciltacabtagene autoleucel (74%). The baseline characteristics of the entire cohort, separated into patients with vs without ASIP, are shown in Table 1 and were similar between groups. The median follow-up for the cohort was 25 months (interquartile range [IQR], 14-43). A total of 20 of the 58 patients (34%) had POD after CAR-T therapy. ASIP was observed in 18 patients (31%) during CAR-T follow-up in the absence of evidence of POD (referred to simply as ASIP; Table 1 ). Of these 18 patients, only 5 (28%) had POD at some point after ASIP occurred. Each patient’s individual time course with ASIP occurrence before POD, best response, as well as infectious/vaccination events is shown in Figure 1 A.

Table 1. Baseline characteristics of the BCMA CAR-T–treated cohort Characteristic Patients with preprogression ASIP All patients (N = 58) No (n = 40) Yes (n = 18) P value ∗ Age, median (IQR), y 64 (57-72) 66 (55-69) .9 65 (56-71) Timing of ASIP occurrence, n (%)  Before progression 0 (0) 18 (100) <.001 18 (75)  After progression 6 (100) 0 (0) 6 (25) CAR-T product, n (%)  Ide-cel 9 (23) 6 (33) .5 15 (26)  Cilta-cel 31 (78) 12 (67) 43 (74) CRS, n (%) 34 (85) 18 (100) .2 52 (90) CRS grade, n (%) .4  Grade I 22 (65) 14 (78) 36 (59)  Grade II 10 (29) 2 (11)  Grade III 2 (5.9) 2 (11) ICANS, n (%) 6 (15) 0 (0) .2 6 (10) Heavy-chain isotype, n (%)  IgA 8 (20) 7 (39) .4 15 (26)  IgG 23 (58) 8 (44) 31 (53)  Light chain only 9 (23) 3 (17) 12 (21) Light-chain isotype, n (%)  κ 26 (65) 11 (61) .8 37 (64)  λ 14 (35) 7 (39) 21 (36) ISS stage at diagnosis, n (%)  I 14 (37) 4 (24) .4 18 (33)  II 13 (34) 9 (53) 22 (40)  III 4 (11) 3 (18) 7 (13)  Unknown 7 (18) 1 (5.9) 8 (15) EMD presence, n (%) 17 (43) 6 (33) .5 23 (40) Response (VGPR or better vs PR or worse), n (%)  PR or worse 10 (26) 0 (0) .022 10 (18)  VGPR or better 29 (74) 18 (100) 47 (82) Response (CR, VGPR, and PR or worse), n (%)  CR 20 (51) 15 (83) .027 35 (61)  VGPR 9 (23) 2 (17) 12 (21)  PR or worse 10 (26) 0 (0) 10 (18) Previous lines of therapy, median (IQR), n 5 (4-6) 4 (4-6) .2 5 (4-6) PFS events, n (%) 15 (38) 5 (28) .5 20 (34) The boldface values emphasize significant values for visualization purposes. cilta-cel, ciltacabtagene autoleucel; CR, complete response; CRS, cytokine release syndrome; EMD, extramedullary disease; ICANS, immune effector cell-associated neurotoxicity syndrome; ide-cel, idecabtagene vicleucel; PR, partial response; ISS, International Staging System. ∗ Pearson χ 2 test, Wilcoxon rank-sum test, and the Fisher exact test as appropriate.

Baseline characteristics of the BCMA CAR-T–treated cohort

The boldface values emphasize significant values for visualization purposes.

cilta-cel, ciltacabtagene autoleucel; CR, complete response; CRS, cytokine release syndrome; EMD, extramedullary disease; ICANS, immune effector cell-associated neurotoxicity syndrome; ide-cel, idecabtagene vicleucel; PR, partial response; ISS, International Staging System.

Pearson χ 2 test, Wilcoxon rank-sum test, and the Fisher exact test as appropriate.

Cohort course after CAR-T infusion, progression free survival with and without landmark analysis. (A) Swimmer plot with individual patient timelines, including timing of immunological events, ASIP isotypes, and best response throughout the post–CAR-T course. The triangles for ASIP start and end may overlap if the ASIP event only appeared on 1 serum protein electrophoresis/serum immunofixation electrophoresis. “Other” for immunological events includes cancer or autoimmune disease diagnoses or the administration of tocilizumab. (B) PFS stratified by ASIP vs no ASIP. (C) PFS stratified by ASIP vs no ASIP with landmark time of 7 months (25th percentile of time to ASIP). CR, complete response; ID, identification number; PR, partial response.

There were 23 ASIP events in the cohort of these 18 patients because a few patients had multiple ASIP events at different time points. Information on the characteristics of these ASIP events can be found in supplemental Table 1 . The median time to ASIP after CAR-T was 9 months (IQR, 7-17). The median duration of ASIP was 1.38 months (range, 1-12). Quantification of ASIP monoclonal proteins revealed that most were too faint to quantify by protein electrophoresis or were only detected by serum immunofixation ( supplemental Table 1 ).

All ASIP cases occurred after achieving the patient’s best response, with 15 patients (83%) achieving complete response and 3 patients (17%) achieving very good partial response (VGPR). Patients with ASIP had deeper responses; all 18 patients with ASIP (100%) achieved VGPR or better, whereas 29 of the 40 patients (74%) without ASIP achieved VGPR or better ( P = .022; Table 1 ). This shows that ASIP may arise in the absence of residual disease, which may enhance the detection of non-MM-related paraproteins.

It has been hypothesized that humoral immune reconstitution could be the cause of ASIP. 5 , 13 Molecular studies on ASIP after ASCT have suggested that ASIP are not produced by myeloma cells but by nonneoplastic plasma cells. 14 Another recent study shows that absolute lymphocyte count recovery after BCMA CAR-T is prognostic and that these expanding lymphocytes are not only CAR-T but also non–CAR-T. 15 In our cohort, when comparing patients with and without ASIP, there was no difference in the median immunoglobulin levels after CAR-T (median immunoglobulin G [IgG], 572 vs 502 mg/dL; P = .4; median immunoglobulin A [IgA], 15 vs 10 mg/dL; P = .7; median immunoglobulin M [IgM], 28 vs 15 mg/dL; P = .13), but a trend exists with patients with ASIP having higher immunoglobulin levels. Patients with ASIP had similar rates of overall hypogammaglobulinemia as those without; however, when divided into immunoglobulin subtypes, patients with ASIP spent a lower proportion of time after CAR-T being lower than the lower limit of normal for IgA and IgM, suggesting that patients with ASIP may have a more robust immune reconstitution after BCMA CAR-T therapy. Importantly, there was no difference in baseline immunoglobulin levels pre–CAR-T therapy between those with and without ASIP ( supplemental Data ).

In patients with ASIP, 12 of the 23 ASIP events (52%) were preceded within 60 days by an immunological event, with 9 infections (64%), 2 vaccinations (14%), and 1 with both an infection and vaccination (7%; supplemental Table 2 ). There was no difference in the proportion of patients who had immune events during follow-up, with no difference in the median number of vaccinations, but a significantly higher median number of infections in patients with ASIP (2 events [IQR, 1-3] vs 4 events [IQR, 2-6]; P = .021; supplemental Table 3 ). Only reported immune events could be identified, and it is possible that patients could have had unreported/minor infections or infections not detected by standard testing. Given the retrospective and self-reported nature of infections, there are limitations to this analysis. Although all patients had monthly follow-up with serum protein electrophoresis/serum immunofixation electrophoresis, patients with more frequent infections could potentially have had higher testing, which may also contribute to increased detection of ASIP. Importantly, based on the results presented here, there was no signal suggesting ASIP is associated with increased infection burden after BCMA CAR-T.

Of the 18 patients in our cohort with ASIP, 6 (33%) had mass spectrometry performed at the time of ASIP. Uniformly, all 6 patients with mass spectrometry (100%), some with multiple samples during ASIP, had negative test results with multiple small clones being found, supporting the conclusion that ASIP may actually be an oligoclonal or polyclonal phenomenon.

PFS showed a statistically significant survival benefit for those with ASIP (not reached [95% confidence interval, 32.2 to not reached] vs 18.4 months [95% confidence interval, 13.0 to not reached]; P = .023; Figure 1 B). However, ASIP itself creates a survival bias because of its late onset after CAR-T therapy. To account for this, a landmark analysis was conducted. When begun at 7 months after CAR-T therapy (the 25th percentile of time to ASIP occurrence), the statistically significant PFS benefit observed becomes insignificant ( Figure 1 C).

In summary, this is the largest reported cohort detailing the occurrence of ASIP after BCMA CAR-T therapy, 16 , 17 , 18 with 31% of the patients experiencing ASIP and 100% of ASIP occurring after achieving a deep response (VGPR or better). To our knowledge, we are the first in this context to show that the PFS benefit observed may be because of a selection bias inherent to ASIP. Another study on ASIP after ASCT conducted a landmark analysis, but survival was still prolonged, indicating that ASIP may have different properties after different therapies. 3 The data presented support the hypothesis that ASIP occur in patients with a deeper response and better humoral immune reconstitution and/or preservation of nonmalignant plasma cells in response to an immunological stimulus. Importantly, there was no signal for increased risk of relapse because of the loss of therapeutic pressure on BCMA + plasma cells or evidence of increased infection burden in patients with ASIP. Future studies are needed to validate our findings and investigate further the immune reconstitution after CAR-T therapy, both humoral and cellular, because this may allow for the identification of patients with increased risk of infection after CAR-T therapy.

Conflict-of-interest disclosure: J.M. reports consulting fees from Johnson & Johnson, Pfizer, and Sanofi. D.B. reports consultancy fees and research funding from Sanofi. R.C. reports receiving consultancy fees from Pfizer, Adaptive Biotechnologies, Sanofi, Alexion, and Janssen; honoraria from Pfizer, Adaptive Biotechnologies, Sanofi, Alexion, and Janssen; and research funding from Genentech and AbbVie. S.L. reports receiving advisory or consulting fees from Pfizer, Janssen, GSK, Sanofi, Bristol Myers Squibb, Karyopharm, and Regeneron; honoraria from Regeneron, Clinical Care Options, Springer Healthcare, PeerView, Medscape, and Aptitude; and research funding from Zentalis and Sanofi. R.R. reports receiving advisory or consulting fees from Incyte, Bayer, Sana Biotechnology, Orca Bio, Gilead Sciences, Quell Biotherapeutics, Autolus, TScan, and Allogene and research funding from Incyte, TCR2, Gilead Sciences, Sanofi, Atara Biotherapeutics, Bristol Myers Squibb, Takeda, AbbVie, Immatics, Genentech, Johnson & Johnson, Synthekine, CareDx, Cabaletta, TScan, and Precision Biosciences. R.N. received honoraria from Sanofi, Merck, Bristol Myers Squibb, Janssen, Takeda, and Amgen; received research funding from Sanofi and Merck; and is a current equity holder at Karyopharm. M.B. reports receiving consultancy fees from Janssen Oncology, Takeda, and Bristol Myers Squibb and honoraria from Janssen Oncology, Karyopharm Therapeutics, Menarini Silicon Biosystems, Takeda, Sanofi, and Bristol Myers Squibb. M.M.S. reports consulting fees from Johnson & Johnson/Legend Biotech. The remaining authors declare no competing financial interests.

Acknowledgments

Acknowledgments: The authors thank all patients who participated in this study and their caregivers.

This work was supported in part by the Weill Cornell Medicine Funds for the Future grant (M.B.), the American Society of Clinical Oncology
Conquer Cancer Young Investigator award (M.M.S. and M.B.), and the American Society of Hematology 66th annual meeting Abstract Achievement Award (C.U.).

Contribution: M.M.S., M.B., J.M, and C.U. designed the study; C.U. and M.M.S. performed data analysis and interpretation and wrote the manuscript; C.U., E.O.-V., C.A.G., and J.E.V.-H. collected the data; J.M., D.B., R.P., R.C., S.L., R.R., and R.N. conceptualized and designed the study; M.M.S. and M.B. collected data, analyzed and interpreted the results, and elaborated on the manuscript; and all authors reviewed and contributed to the final manuscript.

Footnotes

Presented in abstract form at the 21st annual meeting of the International Myeloma Society, Athens, Greece, 26 September 2024 19 ; the 66th annual meeting of the American Society of Hematology, San Diego, CA, 7 December 2024 20 ; and the 22nd annual meeting of the International Myeloma Society, Toronto, ON, 18 September 2025. 21

Data are available from the corresponding author, Mateo Mejia Saldarriaga ( mam9823@med.cornell.edu ), on request.

The full-text version of this article contains a data supplement.

Supplementary Material

Supplemental Methods, Tables, and Reference

References

Associated Data

Supplementary Materials

Supplemental Methods, Tables, and Reference