Impact of interval progression before autologous stem cell transplant in patients with multiple myeloma

Abstract

In transplant-eligible patients who undergo upfront autologous stem cell transplant (ASCT) for multiple myeloma (MM), standard practice is to treat with six to eight cycles of induction therapy followed by high-dose chemotherapy with ASCT. A gap between the end of induction and the day of ASCT exists to allow stem cell mobilization and collection. Despite attempts to limit the length of this interval, we noticed that some patients experience interval progression (IP) of disease between the end of induction therapy and the day of ASCT. We analyzed 408 MM patients who underwent ASCT between 2011 and 2016. The median length of the interval between end of induction and ASCT was 38 days. We observed that 26% of patients in the entire cohort and 23.6% of patients who received induction with bortezomib-lenalidomide-dexamethasone (VRD) experienced IP. These patients deepened their responses with ASCT, independently of induction regimen. In the entire cohort, IP was significantly associated with shorter PFS in the univariable analysis (Hazard Ratio, HR = 1.37, P = 0.022) but not in the multivariable analysis (HR = 1.14, P = 0.44). However, analyzing only patients who received VRD as induction, progression-free survival (PFS) remained inferior in both the univariable (HR = 2.02; P = 0.002) and the multivariable analyses (HR = 1.96; P = 0.01). T cells and natural killer (NK) cells are increasingly studied targets of immunomodulatory therapy, as immune dysfunction is known to occur in patients with MM. Peripheral blood from 35 MM patients were analyzed. At time of ASCT, patients with IP had significantly increased percentages of CD3⁺CD8⁺CD57⁺ CD28^- (P = 0.05) and CD3⁺CD4⁺LAG3⁺ (P = 0.0022) T-cells, as well as less CD56^bright and CD56^dim NK cells bearing activated markers such as CD69, NKG2D, and CD226. These data suggest that IP can impact the length of response to ASCT; therefore, further studies on the management of these patients are needed.

Keywords: disease progression; immune profiling; myeloma; outcome; stem cell transplant (SCT).

Figure 1

Post-induction, pre-transplant, and post-transplant responses in patients with MM. (A) Post-induction responses, pre-transplant responses, and post-transplant responses in non progressors (NP) or patients with interval progression (IP) are shown. Responses are presented as complete responses (CR), very good partial response (VGPR), partial response (PR), and minimal response or stable disease (MR/SD). Statistical test: Fisher exact test. P-values are as such: Post-induction P < 0.001, ***; pre-transplant P < 0.001, ***; post-transplant P < 0.001, ***. (B) The percentage of patients with interval progression (IP) (n = 106) who changed responses between end of induction and pre-transplant is shown. Responses are classified as in (A). Descriptive statistics is used in the analysis.

Figure 2

Progression-free survival (PFS) and overall survival (OS) in non-progressors (NP) and patients with interval progression (IP). (A) Kaplan-Meier survival curves for PFS from day of autologous stem cell transplant (ASCT). Log-rank P = 0.022. (B) Kaplan-Meier survival curves for OS from day of ASCT. Log-rank P = 0.2.

Figure 3

Ad hoc analysis in the cohort of patients who received VRD as induction strategy. (A) Post-induction responses, pre-transplant responses, and post-transplant responses in n = 126 non progressors (NP) or n = 39 patients with interval progression (IP) are shown. Responses are presented as complete responses (CR), very good partial response (VGPR), partial response (PR), and minimal response or stable disease (MR/SD). Statistical test: Fisher exact test. P-values are as such: Post-induction P = 0.01, *; pre-transplant P < 0.001, ***; post-transplant P = 0.01, *. (B) The percentage of patients with interval progression (IP) (n = 106) who changed responses between end of induction and pre-transplant is shown. Responses are classified as in (A) Descriptive statistics is used in the analysis. (C) Kaplan-Meier survival curves for PFS from day of autologous stem cell transplant (ASCT). Log-rank P value = 0.001. (D) Kaplan-Meier survival curves for OS from day of ASCT. Log-rank P value = 0.03.

Figure 4

Increase in specific T cell populations in patients with interval progression (IP) compared with non-progressors (NP). The expression of the different markers is assessed in the peripheral blood of 7 healthy donors (HD), 22 NP, and 13 patients with IP. Percentages of (A) CD3⁺CD8⁺CD57⁺CD28^- T cells in HD, NP, and IP. One-way ANOVA with Bonferroni’s multiple comparison test: NS, HD versus NP; NS, HD versus IP; *, P = 0.05, NP versus IP; ANOVA summary P = 0.05. (B) CD3⁺CD4⁺LAG3⁺ T cells in HD, NP, and IP. One-way ANOVA with Bonferroni’s multiple comparison test: NS, HD versus NP; *, P = 0.017; HD versus IP; **, P = 0.0022, NP versus IP; ANOVA summary P = 0.0016. (C) CD3⁺CD4⁺PD-1⁺CD25^- T cells in HD, NP, and IP. One-way ANOVA with Bonferroni’s multiple comparison test: NS, HD versus NP; **, P = 0.0089, HD versus IP; *, P = 0.0305, NP versus IP; ANOVA summary P = 0.0055.

Figure 5

Accumulation of less mature NK cells in non-progressors (NP) compared with patients with interval progression (IP). (A) Schema of NK cell development with activation markers from NK precursors, to immature CD56^brightCD16^+/- cells to mature CD56^dimCD16⁺ cells, expressing activator markers. The expression of the different markers is assessed in the peripheral blood of 7 healthy donors (HD), 22 NP, and 13 patients with IP. Median with interquartile ranges are shown. Percentages of: (B) CD3^- CD56^bright CD16^-/+ cells in HD, NP, and IP. One-way ANOVA with Bonferroni’s multiple comparison test: NS, HD versus NP; *, P = 0.0125, HD versus IP; NS, P = 0.23, NP versus IP; ANOVA Summary P = 0.0145. (C) CD3^- CD56^bright NKG2D⁺ or CD56^brightCD69⁺ cells in HD, NP, and IP. One-way ANOVA with Bonferroni’s multiple comparison test: NKG2D: NS, HD versus NP; *, P = 0.0454, HD versus IP, *, P = 0.05, NP versus IP, ANOVA Summary P = 0.0452; CD69: NS, HD versus NP; NS, HD versus IP *, P = 0.031, NP versus IP; ANOVA Summary P = 0.13. (D) CD3^-CD56^dim cells in HD, NP, and IP. One-way ANOVA with Bonferroni’s multiple comparison test: NS, P = 0.072, HD versus NP; **, P = 0.0034, HD versus IP; *, P = 0.049, NP versus IP. (E) CD3^-CD56^dimCD16⁺ cells in HD, NP, and IP. One-way ANOVA with Bonferroni’s multiple comparison test: ****, P < 0.0001, HD versus NP; ****, P < 0.0001, HD versus IP; NS, P = 0.99, NP versus IP; ANOVA Summary P < 0.0001. (F) CD3^- CD56^dim CD16⁺ NKG2D⁺ cells in HD, NP, and IP. One-way ANOVA with Bonferroni’s multiple comparison test: NS, HD versus NP; ***, P = 0.0009, HD versus IP; *, P = 0.046, NP versus IP; ANOVA Summary P = 0.0012; CD3^- CD56^dim CD16⁺ CD226⁺ cells in HD, NP, and IP. One-way ANOVA with Bonferroni’s multiple comparison test: ***, P = 0.0005, HD versus NP; ****, P < 0.0001, HD versus IP; *, P = 0.05, NP versus IP; ANOVA Summary P < 0.0001; and CD3^- CD56^dim CD16⁺ CD69⁺ cells in HD, NP, and IP. One-way ANOVA with Bonferroni’s multiple comparison test: **, P = 0.0052, HD versus NP; ****, P < 0.0001, HD versus IP; **, P = 0.0091, NP versus IP; ANOVA Summary P < 0.0001.