Abstract

Clonal hematopoiesis of indeterminate potential (CHIP) is defined by the age-dependent accumulation of somatic leukemia-associated driver mutations in hematopoietic stem cells, in individuals with normal blood counts and with absence of an underlying myeloid neoplasm (MN).1, 2 CHIP is associated with an increased risk of developing MN and an increased all-cause mortality, largely due to cardiovascular disease.3 The presence of CHIP prior to receiving chemotherapy and radiation has been associated with therapy related MN (T-MN), such as myelodysplastic syndromes (MDS) and acute myeloid leukemia.4 Autologous stem cell transplantation (ASCT) is an effective treatment modality for managing higher-risk patients with non-Hodgkin's lymphoma (NHL) and multiple myeloma (MM). In a seminal NHL study, 30% of patients were found to have CHIP at the time of ASCT, with the presence of CHIP being associated with an increased rate of T-MN (10-year cumulative incidence of 14.1% vs 4.3%) and an inferior overall survival (10 year OS 30.4% vs 60.9%).5 In MM, targeted sequencing of 629 patients prior to ASCT detected CHIP in 21.6% of patients, with the presence of CHIP strongly associating with inferior OS (HR 1.34, p = .02) and an inferior progression-free survival (PFS, HR 1.45, p < .001). Interestingly, in this study, adverse CHIP-associations were apparently overcome by lenalidomide maintenance therapy.6 Unlike in NHL, CHIP in MM was not associated with T-MN; while lenalidomide maintenance therapy, independent of the presence or absence of CHIP, was associated with T-MN (p = .047) and second primary malignancies (SPM).6 We carried out this study to assess the prevalence and prognostic impact of CHIP in a relatively uniform cohort of MM patients at the time of ASCT, with all patients going on to receive lenalidomide maintenance therapy. Successive MM patients who consented to have their pre-ASCT bone marrow (BM) sample collected and who underwent first ASCT at Mayo Clinic, followed by lenalidomide maintenance therapy, were included in the study. The BM mononuclear cell DNA from pre-ASCT samples was extracted after excluding CD38/CD138+ (negative selection) plasma cells and then subjected to targeted NGS testing (42-myeloid related genes) by previously described methods.7 All patients were closely followed for the development of T-MN as defined by the 2016 WHO criteria, arterial and venous thromboembolism (VTE) and SPM.8, 9 Response to therapy was assessed using the international myeloma working group (IMWG) consensus criteria 2016.10 Statistical methods are highlighted in the supplemental material. Clonal hematopoiesis was detectable in 23 (23%) of 101 MM patients assessed in the study (Table S1, Figure S1). Clinical characteristics, MM risk stratification, median number of prior therapies, response to therapy, ASCT conditioning regimens, engraftment data, day +100 post ASCT outcomes and median duration of lenalidomide maintenance are outlined in Table 1. Except for a higher median age at MM diagnosis in MM patient with CHIP (p = .002), there were no other significant differences between the two groups (Figure S2). Ten (43.5%) patients in the MM CHIP group and 30 (38.0%) in the MM no CHIP group, received alkylatingagent-based induction therapy prior to ASCT (p = .66). Melphalan 200 mg/m2 conditioning was used in 87.1% of patients (95.7% in the CHIP vs 84.6% in the no CHIP group, p = .45), while the remainder received melphalan 140 mg/m2 conditioning. The median duration of lenalidomide maintenance therapy was 21 months (10–36); 16 months in MM patients with CHIP and 22 months in MM patients without CHIP (p = .76), with the median lenalidomide dose being 15 mg (range 10–15 mg; 10 mg in the CHIP group and 15 mg in the no CHIP group, p = .08). The most frequent CHIP mutations encountered included DNMT3A [52%; median variant allele frequency (VAF) 9.0%, range 2.0%–29.2%], TET2 (26%; median VAF 3.0%, range 2.0%–4.8%), followed by TP53, PPM1D and BRAF (10.0% each), respectively (Figures 1(A) and S1). Sixteen patients (69.6%) had one mutation, while seven (30.4%) had >1 mutation and two patients had four mutations each (Figure 1(B)). Eight (66.6%) of 12 patients with DNMT3A mutations had nonsynonymous missense mutations, while four had deletion variants: with no patient harboring the commonly mutated DNMT3A R882 hot spot. There were no statistically significant differences in CHIP mutation distribution, including TP53 and PPM1D mutations, between MM CHIP patients that received alkylating-agent based induction therapy prior to ASCT, vs MM CHIP patients that did not (Figure 1(C)). At last follow up, 70 (69.3%) relapses after ASCT and 41 (40.6%) deaths were documented. Twenty-nine patients (28.7%) were on salvage therapy whereas 13 (12.9%) were on lenalidomide maintenance and 14 (13.8%) were on observation alone, with no statistically significant differences between MM patients with CHIP vs MM without CHIP. Rates of VTE were similar between MM patients with CHIP (30%) and those without CHIP (24%), with similar rates of provoked thromboses (33% vs 29%, p = .4). Thromboses were diagnosed in typical locations in individuals with CHIP including seven lower extremity deep vein thromboses (DVT) and two pulmonary emboli (PE). In contrast, a variety of VTE locations were diagnosed in those without CHIP, including 10 lower extremity DVT, three PE, one DVT with PE, two upper extremity DVT and one portal vein thrombosis. There was a distinction in VTE timing with regards to lenalidomide use between MM patients with CHIP and those without CHIP. For MM patients with CHIP, 2/9 (22.2%) VTE occurred while on lenalidomide, 2/9 (22.2%) occurred prior to lenalidomide and 4/9 (44.4%) occurred at least 3 months after discontinuation of lenalidomide therapy. For MM patients without CHIP, majority (13/17, 76.5%) of VTE occurred while on lenalidomide, with a minority (11%) occurring either before or at least 3 months after discontinuation of lenalidomide. While lenalidomide is a known risk factor for thrombosis in MM, the fact that 44% of VTE in MM with CHIP occurred >3 months after discontinuing lenalidomide, suggests that CHIP might increase VTE risk in this setting (P = .04). Median OS from the time of diagnosis of the entire cohort was 124.6 months (95%CI 97.5-N/A months) with a corresponding 5 year OS of 82.0% (95% CI 74.8%–89.9%). There was no difference in median OS between MM with CHIP vs those without CHIP (100.2 months; 95%CI 76.2-NA months vs 135.6 months; 95%CI 106.3-N/A months, p = .27) (Figure 1(D)), including assessments with individual CHIP-mutations. The median EFS after ASCT was 36.4 months (95%CI 30.5–48.5 months), with there being no difference in median PFS between MM CHIP (36.4 months, 95%CI 24.1–58.5 months) patients vs MM patients without CHIP (36.4 months, 95%CI 29.9–52.4 months) (p = .34) (Figure 1(E)), including TP53 mutations (Figure S3). There were also no differences between the two groups with regards to non-relapse mortality (Figure S4), cumulative incidence of relapse after ASCT (Figure S5) and time to next treatment. Nineteen (18.8%) SPM were documented, 7 (30.4%) in MM CHIP group vs 12 (15.3%) in the MM no CHIP group (Figure 1(F), p = .13), with corresponding 5-year cumulative incidence rates of 22% and 13%, respectively. These SPM included five (4.9%) hematological malignancies (two in MM CHIP vs three in MM no CHIP), seven (6.9%) skin cancers (three in MM CHIP vs four in MM no CHIP) and seven visceral malignancies (two in MM CHIP vs five in MM no CHIP) (Table S2, Figure S6). The two MM CHIP patients who developed T-MN/MDS had TP53 and PPM1D mutations, respectively. In the MM no CHIP group, there was one patient with B-acute lymphoblastic leukemia and two patients with T-MDS with monosomal karyotypes. In summary, we define the CHIP landscape in MM patients at the time of ASCT, with mutations in epigenetic regulator genes being most common (57%), followed by tumor suppressor genes (17%). Unlike in NHL, presence of CHIP at time of ASCT in MM did not impact OS, PFS and incidence of T-MN; a finding potentially attributable to immunomodulatory properties of lenalidomide, or to the use of maintenance therapy in general.6, 11 While the presence of CHIP and lenalidomide therapy have individually been associated with increased risk of thromboses,1-3, 9 we did not see synergy in MM patients with CHIP, although the timing and patterns of thromboses suggest that CHIP might negatively influence thrombotic risk. While SPM have been well described with lenalidomide maintenance therapy,9 we did not see any differences in SPM and hematological malignancies between the two groups. The findings of this study independently confirm a prior observation on the potential ability of lenalidomide maintenance to mitigate the expected adverse effects of CHIP on OS and PFS in MM patients' post-ASCT6; an important consideration given that approximately 13 000 MM patients undergo ASCT in the US annually, with lenalidomide maintenance considered as standard of care.12, 13 Given the smaller sample size and the inherent flaws of a retrospective analysis, future clinical trials evaluating therapies in MM patients' post-ASCT should consider accounting for the presence of CHIP and its impact on outcomes. The authors would like to acknowledge the “Henry Predolin Leukemia Foundation”, Mayo Clinic, Rochester, MN, USA. Mrinal Patnaik has served on the advisory board of Kura Oncology. A Keith Stewart has served on the advisory board for Celgene. Rafael Fonseca has the following disclosures: Consulting: Amgen, BMS, Celgene, Takeda, Bayer, Janssen, Novartis, Pharmacyclics, Sanofi, Karyopharm, Merck, Juno, Kite, Aduro, OncoTracker, Oncopeptides, GSK, AbbVie. Scientific Advisory Board: Adaptive Biotechnologies, Caris Life Sciences and OncoTracker. Gene mutations annotated in the study have been provided in the supplementary material. Raw sequencing data can be made available on request. Appendix S1 Supporting Information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call