Prognostic relevance of clonal hematopoiesis in myeloid neoplastic transformation in patients with follicular lymphoma treated with radioimmunotherapy

Abstract

While novel radioisotope therapies continue to advance cancer care, reports of therapy-related myeloid neoplasms (t-MN) have generated concern. The prevalence and role of clonal hematopoiesis (CH) in this process remain to be defined. We hypothesized that: (i) CH is prevalent in relapsed follicular lymphoma and is associated with t-MN transformation, and (ii) radiation in the form of radioimmunotherapy (RIT) plays a role in clonal progression. In this retrospective cohort study, we evaluated the prevalence and prognostic impact of CH on clinical outcomes in 58 heavily pre-treated follicular lymphoma patients who received RIT. Patients had been given a median of four lines of therapy before RIT. The prevalence of CH prior to RIT was 46%, while it was 67% (P=0.15) during the course of RIT and subsequent therapies in the paired samples. Fourteen (24%) patients developed t-MN. Patients with t-MN had a higher variant allele fraction (38% vs. 15%; P=0.02) and clonal complexity (P=0.03) than those without. The spectrum of CH differed from that in age-related CH, with a high prevalence of DNA damage repair and response pathway mutations, absence of spliceosome mutations, and a paucity of signaling mutations. While there were no clear clinical associations between RIT and t-MN, or overall survival, patients with t-MN had a higher mutant clonal burden, along with extensive chromosomal abnormalities (median survival, afer t-MN diagnosis, 0.9 months). The baseline prevalence of CH was high, with an increase in prevalence on exposure to RIT and subsequent therapies. The high rates of t-MN with marked clonal complexities and extensive chromosomal damage underscore the importance of better identifying and studying genotoxic stressors accentuated by therapeutic modalities.

Figure 1.

Swimmer plot for group 1 patients (patients with paired samples, with radioimmunotherapy administered between samples). PT_9, 30, 31, and 32 received a total of 6, 6, 3, and 5 different therapies, respectively. RIT: radioisotope therapy; PT: patient; NA: not available; CH: clonal hematopoiesis; TMN: treatment-related myeloid malignancy.

Figure 2.

Swimmer plot for group 2 patients (patients with one or more samples taken after radioimmunotherapy). Symbols and abbreviations as in Figure 1.

Figure 3.

Swimmer plot for group 3 patients (patients with samples taken only prior to radioimmunotherapy). Symbols and abbreviations as in Figure 1.

Figure 4.

Mutation prevalence and co-mutation status. The light blue color indicates the number of patients with the specific mutation, and the light pink color indicates the number of patients with other mutations.

Figure 5.

Co-mutation plot showing mutations present in 35 patients with somatic mutation(s). Each column represents a single patient. The top row denotes the translational effect and mutations per megabase. The bar graph on the left designates the prevalence of mutations (count for once regardless of the prevalence/patient). The mutation subtypes are represented by colors, red indicates nonsense, light-blue indicates missense; yellow indicates frameshift; gray indicates splice effect; purple indicates duplication, pink indicates deletion. The bar at the bottom designates the proportion of the mutation subtypes for each patient. Proportions are from 0 (bottom) to 25%, 50%, 75%, and 100% (top). MB: megabase.

Figure 6.

Changes in variant allele fraction in patients with paired samples in group 1, with two paired serial samples in subgroup 2a; and with two paired serial samples in subgroup 3a. VAF: variant allele fraction; RIT: radioimmunotherapy.