Assessment of the IPSS-M in Chronic Myelomonocytic Leukemia

Background

The CPSS-Mol scoring system is specific to chronic myelomonocytic leukemia (CMML) and identifies 4 risk groups by integrating various risk factors, including the percentage of bone marrow (BM) blasts, white blood cell count (WBC), transfusion dependency, the presence of cytogenetic abnormalities and mutations in ASXL1, NRAS, RUNX1 and SETBP1. Recently, the IPSS-M score has been developed for patients with myelodysplastic syndrome (MDS), classifying patients into 6 risk groups. This index considers the presence of mutations in 31 genes, among other variables. IPSS-M was developed from a series of 2,957 patients, primarily MDS, but also included 272 cases with the myelodysplastic variant of CMML (MD-CMML). The aim of this study was to evaluate the utility of the IPSS-M in CMML, compare both scoring systems, and explore their applicability in the myelodysplastic (MD-CMML) and myeloproliferative (MP-CMML) variants of the disease.

Methods

Clinical and molecular data were collected retrospectively in a series of patients from the Spanish Registry of MDS and the Munich Leukemia Laboratory, diagnosed with CMML according to the 2022 WHO classification. CPSS-Mol and IPSS-M scores were calculated. Overall survival (OS) was assessed using the Kaplan-Meier method and comparisons were performed with the log-rank test. Cumulative incidence of progression to acute myeloid leukemia (AML) was analyzed by competing risks. The discriminative ability of the CPSS-Mol and IPSS-M models was evaluated using the concordance index (Harrell's c-index).

Results

A total of 449 cases of CMML were included with a median follow-up of 1.91 (0.07-15.35) years. Median age at diagnosis was 73 years old (range 27-95) and 69.0% (n=310) were men. According to 2022 WHO, 80.4% (n=361) of patients were classified as CMML-1 and 19.6% (n=88) as CMML-2. As per the FAB classification, 66.6% (n=299) of cases were classified as MD-CMML and 33.4% (n=150) as MP-CMML. Regarding peripheral blood counts, median values were: Hb 11.8g/dL (4.9 – 17.8), WBC 9.00x10⁹/L (2.40 – 93.67), ANC 4.11x10⁹/L (0.10 – 54.73), monocytes 1.80x10⁹/L (0.52 – 36.60), monocyte percentage 21.9% (10.0 – 65.0) and platelets 130 x10⁹/L (6 – 1377). Median percentage of blasts and promonocytes in BM was 4.0% (0.0 -19.0). Overall, median OS was 4.52 years (3.99-6.05) and 17.1% (n=77) of patients progressed to AML, with an AML incidence at 5 years of 21.0% (17.0%-26.0%).

According to CMML specific cytogenetic risk classification, 87.1% (n=391) were classified as low risk (376 cases with normal karyotype and 15 with isolated –Y), 6.0% (n=27) as intermediate risk, and 6.9% (n=31) as high risk (14 cases with +8, 11 with -7/del(7q), and 6 with complex karyotype). Most patients (93.5%, n=420) presented with at least one mutation according to targeted NGS study. Most frequently (>10%) mutated genes were TET2 (63.9%), ASXL1 (37.0%), SRSF2 (36.7%), RUNX1 (16.3%), CBL (13.1%), NRAS (12.0%) and ZRSR2 (10.2%).

After applying the CPSS-Mol, patients were classified into 4 risk groups with significantly different OS and AML incidence at 3 years (p<0.001): low (OS 79%, AML 4%), intermediate-1 (OS 76%, AML 7%), intermediate-2 (OS 56%, AML 19%), and high (OS 40%, AML 34%). Using the IPSS-M, they were categorized into 6 groups with significantly different OS and AML incidence at 3 years (p<0.001): very low (OS 91%, AML 2%), low (OS 68%, AML 13%), moderate-low (OS 68%, AML 20%), moderate-high (OS 50%, AML 15%), high (OS 42%, AML 24%), and very high (OS 24%, AML 43%).

After grouping CPSS-Mol into 2 risk groups and IPSS-M into 3 risk groups, both scoring systems showed a similar discriminative capacity, with the IPSS-M being superior; with a c-index for OS of 0.608 for CPSS-Mol and 0.647 for IPSS-M; and a c-index for AML progression risk of 0.576 for CPSS-Mol and 0.614 for IPSS-M. Then we assessed the utility of the scores in the MD and MP variants. Both were effective in MD-CMML (c-index for OS of 0.616 for CPSS-Mol and 0.663 for IPSS-M; c-index for AML incidence of 0.555 for CPSS-Mol and 0.623 for IPSS-M), and in MP-CMML (c-index for OS of 0.509 for CPSS-Mol and 0.573 for IPSS-M; c-index for AML incidence could not confidently be assessed due to the low number of events in the low risk groups). Both scores showed better discriminative capacity among MD-CMML.

Conclusions

The IPSS-M is applicable in patients with CMML, in both MD-CMML and MP-CMML, and shows a superior discriminative capacity compared to CPSS-Mol.