To the Editor:

Multiple myeloma (MM) is a genetically heterogenous plasma cell malignancy that is characterized by the accrual of subclonal genetic changes as the disease advances [1]. These subclonal alterations contribute to the development of therapeutic resistance and despite the improvement in the overall survival (OS), MM remains incurable [2]. Proteasome inhibitors (PIs) and immunomodulatory drugs (IMiDs) had been the backbone of therapeutic regimens in MM until recently [3]. Monoclonal antibodies (mAbs) against CD38 and SLAMF7 can improve outcomes for patients previously treated with PIs and IMiDs [4]. However, patients with diseases that are refractory to PIs, IMiDs, and mAbs have a poor prognosis and are commonly referred to as having a triple class refractory (TCR) disease [5]. Currently, there is no established standard of care for TCR MM with promising treatment options including selinexor, bispecific antibodies, and chimeric antigen receptor T(CAR-T) cell therapies [4, 6, 7]. In this study, we report the outcomes of patients with triple refractory MM and response to therapy post TCR status. Consecutive patients with MM diagnosed between 01/01/2013 and 12/31/2018 and seen at Mayo Clinic, Rochester, MN were included in this study. Refractoriness to therapy was defined as non-responsiveness (failure to achieve at least a minimal response) to therapy or progression on or within 60 days of discontinuation of therapy [8]. Patients meeting the definition for refractory disease for at least one PI, IMiD, and monoclonal antibody (daratumumab, isatuximab or elotuzumab) were considered to have TCR disease. The mSMART 3.0 (www.msmart.org) classification was used to characterize cytogenetic risk using interphase fluorescence in situ hybridization (FISH), with high-risk cytogenetic features including t(4;14), t(14;16), t(14;20), deletion 17p, TP53 mutation and 1q gain. Patients with FISH data available within 6 months of TCR date was utilized for comparison with baseline FISH cytogenetic data at diagnosis of MM. The international myeloma working group (IMWG) criteria were used for response assessment.

Between 01/01/2013 and 12/31/2018, out of 1773 patients with MM, a total of 249 patients (14%) were identified to have TCR MM. The median follow-up from diagnosis of MM till the date of the last follow-up was 6.8 years (95% CI: 6.3–7.3 years). The median follow-up from TCR MM till the last follow-up was 1.9 years (95% CI: 1.6–2.4). Characteristics of the cohort are depicted in Table 1. Cytogenetic data were available in 208 patients at diagnosis; 116 (56%) patients harbored high-risk cytogenetics. At TCR status, 110 (44%) patients had FISH data available; 95 patients (87%) had high-risk cytogenetics (Table 1). Among these 110 patients, 91 patients also had FISH data available from the time of diagnosis for comparison. Thirty-one percent (28/91) patients did not demonstrate any new cytogenetic abnormality at TCR status; 31% (28/91) had a new 1q duplication, 25% (23/91) had a new 17p deletion, 24% (22/91) had a new 8q24 deletion, 19 (21%) patients had a near tetraploid clone, 3% (3/91) patients had a new deletion 1p.

Table 1 a. Clinical characteristics at diagnosis and TCR status, b. Refractoriness to proteasome inhibitors, immunomodulatory imide drugs and monoclonal antibodies, c. First salvage treatment post triple refractory status.
Full size table

The median number of lines of therapy for the entire cohort were 8 (range 1–17); the patients were exposed to a median of 5 (range 1–12) lines of therapy prior to TCR status. Details of refractoriness to PIs, IMiDs, and mAbs are outlined in 1b. For the entire cohort, 21 (8%) patients were treated with chimeric antigen receptor T (CAR-T) cell therapy, 34 (14%) received a B-cell maturation antigen (BCMA)-directed antibody (23 received an antibody-drug conjugate and 11 received a T-cell redirecting antibody), 20 (8%) received selinexor-based therapy and 41 (16%) were treated with venetoclax-based combination therapy. With regard to transplant therapies from the diagnosis, 181 (73%) patients underwent 1 autologous stem cell transplant (ASCT), 24 (10%) underwent 2 ASCTs, including tandem (7/24) or for recurrence (17/24), 9 (4%) patients received an allogenic transplant in addition to ASCT while 44 (18%) patients did not receive any form of transplant.

The median time to development of triple refractory disease was 2.9 years (95% CI: 2.6–3.2 years) from diagnosis of MM. The time to TCR disease was shorter for high-risk cytogenetics at 2.3 years (95% CI: 2–2.7) compared to 3.4 years (95% CI: 3.1–4.1) with a hazard ratio (HR) of 1.8 (95% CI: 1.4–2.4) p < 0.0001], Fig. 1A. The depth of response to induction therapy did not significantly impact the time to development of TCR disease (Supplementary Fig. 1). At data cutoff, 160 (64%) patients had died. The median OS for the cohort from diagnosis was 4.6 years (95% CI: 4.2–5.2). High-risk cytogenetics were associated with a significantly inferior OS from diagnosis of MM [median 3.8 years (95% CI: 3.1–4.4)] compared to standard risk cytogenetics [median 5.9 years (95% CI: 5.2–7); p < 0.001], Fig. 1B. The median OS from TCR disease was 1 year (95% CI: 0.9–1.3), Supplementary Fig. 2. High-risk cytogenetics at diagnosis were associated with an inferior OS even from TCR disease with a median OS of 1 year (95% CI: 0.8–1.5) compared to 1.3 years (95% CI: 0.9–2.9) for standard risk cytogenetics [HR 1.44 (95% CI: 1.01–2.05) p = 0.04] (Supplementary Fig. 3). High-risk cytogenetics at diagnosis were independently associated with inferior OS after adjusting for ISS stage at diagnosis and age for TCR disease [HR 1.6 (95% CI: 1.1–2.3); p = 0.02]. For patients with time to TCR status of >3 years, the median OS from TCR status was 1.5 years (95% CI: 1.1–2.3) compared to a median of 0.8 years (95% CI: 0.5–1) for patients with a time-to-TCR status of ≤3 years [HR 0.54, 95% CI: 0.4–0.75; p < 0.001], (Supplementary Fig. 4). Out of 249 patients, 227 had details available for subsequent therapy (Table 1). The median PFS for the initial therapy post TCR MM was 4 months (95% CI: 3.5–5 months), Supplementary Fig. 5. Response data were available in 211 out of 227 patients for first therapy post TCR disease. The objective response rate (PR or better) with first therapy for TCR disease was 45% (95/211). Patients achieving a deeper response [≥partial response (PR)] were noted to have a significantly improved PFS with the initial therapy post TCR disease [median PFS 8.2 months (95% CI: 7.5–12.7)] compared to patients not achieving a PR [median PFS 2 months (95% CI: 1.6–2.7); HR 0.2 (95% CI: 0.13–0.27); p < 0.0001], Fig. 1C. Depth of response (≥PR) was associated with an improved PFS [HR 0.23 (95% CI: 0.16–0.34); p < 0.0001] after adjusting for age at TCR status, ISS stage and cytogenetic risk categories at diagnosis. High-risk cytogenetics at diagnosis were also associated with an inferior PFS for initial therapy after TCR status after adjusting for age, depth of response (≥PR), and ISS stage [HR 1.5 (95% CI: 1.01–2.1), p = 0.04]. The comparison of PFS and response rates for the initial treatment regimens post TCR is depicted in Supplementary Table 1. Out of 249 patients, 102 had penta-refractory disease (refractory to bortezomib, carfilzomib, lenalidomide, pomalidomide, and daratumumab). The median PFS with the next line of therapy for penta-refractory disease was 3.2 months (95% CI: 2.8–4.7) and median OS was 7.1 months (95% CI: 6–13.5).

Fig. 1
figure 1

A Time to Triple Refractory Status: Patients with high-risk cytogenetic were associated with a shorter time to triple class refractory status [median 2.3 years (95% CI: 2–2.7)] compared to 3.4 years (95% CI: 3.1–4.1) for standard risk cytogenetics with a hazard ratio (HR) of 1.8 (95% CI: 1.4–2.4) p < 0.0001]; B Overall Survival by Cytogenetic risk-group: Among TCR patients, high-risk cytogenetic at diagnosis was associated with an inferior overall survival from diagnosis of MM [median 3.8 years (95% CI: 3.1–4.4)] compared to standard risk cytogenetics [median 5.9 years (95% CI: 5.2–7); p < 0.001]; C PFS based on Response to Therapy at TCR status: A deeper response [Partial Response (PR) or better] to initial therapy post TCR status is associated with a significantly improved progression-free survival [median PFS 8.2 months (95% CI: 7.5–12.7)] compared to patients not achieving a PR [median PFS 2 months (95% CI: 1.6–2.7); HR 0.2 (95% CI: 0.13–0.27); p < 0.0001].

Full size image

The incorporation of PIs and IMiDs in frontline therapy of MM has led to significant improvement in survival [9]. More recently, quadruple drug combinations incorporating mAbs, PIs, and IMiDs in frontline therapy, have demonstrated improvement in depth of response as well as progression-free survival [10, 11]. Unfortunately, relapse on treatment with even these quadruple combination therapies is common and this poses a unique problem of encountering these triple refractory patients (refractory to PIs, IMiDs and mAbs) earlier in the disease course. Outcomes of TCR patients is poor, but their clinical course has not been described well. Our study population had a median OS of 1 year from triple refractory status and the median PFS with the initial therapy used for triple refractory patients was 4 months. Selinexor, an exportin inhibitor that has been studied in a triple-class refractory setting, also demonstrated a comparable median PFS of 3.7 months [12]. Another retrospective study compared selinexor versus any other MM-directed therapy for TCR MM demonstrated an OS of 10.4 vs 6.9 months (p = 0.043), respectively [13]. The locoMMotion study included 225 patients with RRMM receiving ≥3 lines of therapy (73% TCR) with an ORR of 20% for salvage therapy; the PFS and OS data was not mature at the time of presentation [14]. The heterogenous nature of treatments utilized as initial therapies for triple refractory disease makes it difficult to have a meaningful intra-group comparison. With the caveats of comparison of small groups, our study demonstrates that patients treated with PI and IMiD combinations as well as BCMA-directed therapies performed better compared to alkylator therapy-based combinations in TCR disease. Nonetheless, the poor PFS and OS from triple refractory status underline the importance of identifying newer efficacious and tolerable therapy options for this group of patients. Newer targets, especially CAR-T and bispecific antibodies, represent promising treatment options for triple refractory disease. [6] A recent study looking at outcomes of the BCMA CAR-T idecabtagene-vicleucel (ide-cel, bb2121) demonstrated a PFS of 11.6 months and OS of 20.2 months in patients with TCR MM [15]. Our cohort had 20% patients treated with a BCMA-directed therapy (23 patients treated with BCMA ADC, 11 with BCMA T-cell redirecting antibody and 21 with BCMA CAR-T therapy). This small proportion of patients exposed to BCMA-directed therapies limits our insight into their impact in our cohort of TCR patients. Our cohort of patients was a heavily pretreated one with a median of 8 lines of therapy with a 15% of patients being refractory to all three PI (bortezomib, carfilzomib, ixazomib), 55% patients refractory to both lenalidomide and pomalidomide and 19% patients refractory to both daratumumab and elotuzumab.

The depth of response and high-risk cytogenetics at diagnosis still represent important prognostic makers even this late in the disease stage. A rapid development of TCR disease is associated with poor outcomes, likely reflecting an unstable genome. The study period included patients diagnosed between 2013 and 2018, making it relevant to the current treatment strategies in MM. A high proportion of patients (64%) in our cohort had died at data cutoff suggesting an adequate follow-up time, adding further strength to the findings. This study comes with the limitations of a retrospective analysis including missing cytogenetic risk classification at diagnosis in ~15% of patients as well absence of treatment data post TCR status in ~10% of patients. Treatment paradigms in MM are evolving rapidly with newer therapies being incorporated into the treatment of relapsed/refractory disease and this study provides valuable benchmark for outcomes in a clinically important setting where prospective data are currently limited.