Real‐world data of chronic myelomonocytic leukemia: A chinese single‐center retrospective study

Abstract Chronic myelomonocytic leukemia (CMML) is a rare disease of elderly people characterized by the presence of sustained peripheral blood monocytosis, overlapping features of myeloproliferation, and myelodysplasia. We present a large retrospective study of 156 CMML patients in China. Mean age at diagnosis was 68 years old (range 23‐91). According to the CMML‐specific prognostic scoring system (CPSS), 10 patients (8.3%) were low risk, 27 patients (22.5%) were intermediate‐1 risk, 72 patients (60%) were intermediate‐2 risk, and 11 patients (9.2%) were high risk. A total of 90 patients (57.7%) received hypomethylating agents (HMAs) treatment, 19 patients (12.2%) received chemotherapy and 47 patients (30.1%) received the best supportive care. Seventeen patients (10.9%) underwent allogeneic hematopoietic stem cell transplantation (allo‐SCT) after HMAs treatment or chemotherapy. With a median follow‐up of 35.3 months, overall response rate (ORR) was 69.5% in the HMAs ± chemotherapy group, 79.5% in the HMAs monotherapy group, 60.0% in the HMAs + chemotherapy group, and 37.5% in the chemotherapy group. HMAs monotherapy group had prolonged OS compared with the chemotherapy group (23.57 months vs. 11.73 months; p = 0.035). Patients who achieved ORR had prolonged OS (25.83 months vs. 8.00 months; p < 0.001) and LFS (20.53 months vs. 6.80 months; p < 0.001) compared with those not achieved ORR in the HMA ± chemotherapy group. By univariate analysis, only higher hemoglobulin (≥80 g/L) and lower serum LDH levels (<300 U/L) predicted for better OS and LFS. By multivariate analysis, only Hb ≥ 80 g/L predicted for prolonged OS, Hb ≥ 80 g/L, and monocytes < 3 × 109/L predicted for prolonged LFS. In summary, our study highlights the benefit of HMAs therapy in CMML, but we still need to develop novel therapeutics to achieve better outcomes.


| INTRODUCTION
Chronic myelomonocytic leukemia (CMML) is a clonal hematological malignancy characterized by peripheral blood monocytosis (≥1 × 10 9 /L, with monocytes ≥ 10% of the total white blood cells), and features of both myelodysplastic syndromes (MDS) and myeloproliferative neoplasms (MPN). 1 CMML is a relatively rare disease with an approximated incidence of 1-4 cases per million every year in western countries and a strong male preponderance. It is mainly diagnosed in older patients, with a median age of 71-74 years at diagnosis. [2][3][4][5] The prognosis of CMML is poor with median overall survival ranging from 12 to 29 months and 14-29% progression probability to acute myeloid leukemia (AML). [6][7][8][9] On the basis of WBC count, the French-American-British (FAB) classification divided CMML into two subtypes: myelodysplastic CMML (MD-CMML) with WBC counts <13 × 10 9 /L and myeloproliferative CMML (MP-CMML) with WBC counts ≥13 × 10 9 /L. 10 Furthermore, 2016 WHO classification identifies three categories of CMML based on blasts of peripheral blood and bone marrow: CMML-0 for patients with <2% blast in PB and <5% blast in BM; CMML-1 for patients with 2-4% blast in PB and/or 5-9% blast in BM; and CMML-2 for patients with 5-19% in PB, 10-19% in BM and/or presence of Auer rods. 1 CMML is a highly heterogeneous disease with a distinctive biological diversity ranging from morphological changes, clinical manifestations, cytogenetic abnormalities, molecular biology and treatment responses. 11 About 20-30% of CMML patients have cytogenetic abnormalities, such as −Y, +8, del(20q), +21, −7, del(7q) and complex karyotype. [12][13][14] Most CMML patients (>=90%) have gene mutations, frequent mutations including TET2 (~60%), SRSF2 (~50%), ASXL1 (~40%), and RAS (~30%). 15 Until now, there is no agreement on the optimal therapy for CMML because of the high heterogeneity of the disease. Treatment modalities for CMML involved allogeneic hematopoietic stem cell transplantation (allo-SCT), hypomethylating agents (HMAs), cytotoxic chemotherapy, and supportive care. 1 Although allo-SCT is the only available curative therapy, only a few patients are eligible due to advanced age and comorbidities. [16][17][18] In the unfit patients for allo-SCT, cytotoxic chemotherapy leads to low response rates and short response duration. 15 Hydroxyurea is best suited for cytoreduction. HMAs such as decitabine and 5-azacitidine have exhibited some effectiveness in postponing disease progression and were approved for the treatment of CMML. 19,20 Recently, several large retrospective studies demonstrated the overall response rates (ORR) of HMAs treatment ranged from 40% to 50% and complete remission rates (CR) were <20%. 21 As there are limited data on CMML in China, even in Asia, we conducted a retrospective study to evaluate the variables that can influence the response rates. We analyzed the patients who received HMAs with or without chemotherapy as compared to the patients who received chemotherapy alone.

| Patients
We reviewed all the patients with a diagnosis of CMML in our hospital during a period from January 2003 to June 2019. One hundred and fifty-six patients with peripheral monocytes >1 × 10 9 /L and bone marrow blast <19% were enrolled according to the 2016 WHO definition of CMML. 1 Splenomegaly, hepatomegaly, and lymphadenectasis were defined as clinical or radiological enlargement. Bone marrow blasts contained promonocytes, myeloblasts, and agranular blasts. CMMLspecific prognostic scoring system (CPSS) was served to evaluate the prognosis of CMML patients. 22 This study was approved by the Ethics Committee of the First Affiliated Hospital, College of Medicine, Zhejiang University, following principles of the Declaration of Helsinki.

| Cytogenetic analysis
Cytogenetic analysis was performed in the Institute of Hematology in our hospital following standard protocols. In accordance with CMML-specific cytogenetic risk classification, cytogenetic risk was categorized into three groups: low risk (normal karyotype or sole −Y), intermediate risk (all other abnormalities not in the high or low-risk groups), and high risk (+8, abnormalities of chromosome 7, or complex karyotype). 13

| Treatment response
The Modified International Working Group (IWG 2006) response criteria 23 were applied to evaluate treatment response.

MA et Al.
Categories of response contained complete remission (CR), partial remission (PR), marrow CR (mCR), hematologic improvement (HI), stable disease (SD), and treatment failure. The overall response rate (ORR) was the sum of CR, PR, and mCR or HI.

| Follow-up
The last follow-up was performed on June 15th 2019. The median follow-up period was 35.3 months (95%CI:14.44-56. 16). The overall survival (OS) was calculated as the period from the day of diagnosis to the day of death regardless of any cause or last contact. Leukemia-free survival (LFS) was calculated from the day of diagnosis to the day of leukemia transformation or death, or last contact. The patients that underwent allo-SCT were examined on the day of transplantation.

| Statistical analysis
The SPSS 22.0 software (SPSS Inc.; Chicago, IL, USA) was used to conduct statistical analysis. Mann-Whitney U test was used to compare the baseline characteristics. Categorical variables were analyzed with the Chi-square test. The analysis for predictors of HMAs response was carried out by means of logistic regression. P values less than 0.05 were regarded as statistically significant. OS and LFS curves were constructed by the Kaplan-Meier method and compared by the log-rank test. Factors associated with OS or LFS were analyzed first with univariate analysis, then followed by multivariate Cox proportional hazard regression.

| Treatment modalities
Among the patients who were treated with HMAs, 81 patients treated with decitabine and 9 patients treated with azacitidine ( Table 1). The median number of cycles of HMAs therapy was four cycles (range 1-17). The median number of cycles of chemotherapy was two cycles (range 1-4). Seventeen patients (10.9%) underwent allo-SCT after HMA treatment or chemotherapy. Six patients (35.3%) died after transplantation: three patients died of SCT related complications in the early stage, three patients transformed to acute myeloid leukemia and died of salvage therapy failure.
Second, we compared treatment response between the HMAs monotherapy group and chemotherapy group (Table 4). ORR was 79.5%, including 5 CR (8.1%), 36 mCR/ HI (58.1%), and 1 PR (1.6%) in HMAs monotherapy group. In contrast, ORR was 37.5% (6 patients achieved mCR/HI and no patients achieved CR or PR) in the chemotherapy group. There was a noteworthy difference in ORR between the two groups (p = 0.027).

| Survival
The median follow-up period was 35.3 months (95% CI: 14.44-56.16) and nine cases were lost to follow-up. Median OS was 23.3 months (95% CI: 20.23--26.37) and median LFS was 19.4 months (95% CI: 13.33-25.47). Twenty-seven patients (16.7%) developed a transformation into acute myelocytic leukemia, verified by bone marrow aspiration and flow cytometry. OS was not significantly different between the two groups (19.37 months in HMAs ± chemotherapy group vs. 11.73 months in the chemotherapy group, p = 0.104) ( Figure 1A), but the HMAs ± chemotherapy group tended to have prolonged OS. There was no apparent difference between the two groups in LFS (14.0 months in HMAs ± chemotherapy group vs. 9.40 months in the chemotherapy group, p = 0.193) ( Figure 1B).
Patients who achieved ORR had prolonged OS (25.83 months vs. 8.00 months; p < 0.001) ( Figure 2A) and LFS (20.53 months vs. 6.80 months; p < 0.001) than those not achieved ORR in HMAs ± chemotherapy group ( Figure 2B). Similarly, patients who achieved ORR had prolonged OS (26.27 months vs. 8.73 months; p < 0.001) ( Figure 2E) and LFS (25.1 months vs. 8.73 months; p < 0.001) ( Figure 2F) than those not achieved ORR in HMAs monotherapy group. Patients who achieved ORR also had prolonged OS (19.40 months vs. 6.35 months; p = 0.001) ( Figure 2G) and LFS (11.03 months vs. 2.14 months; p = 0.030) ( Figure 2H) than those not achieved ORR in the HMAs + chemotherapy group. Conversely, the OS of patients in the chemotherapy group had no relationship with whether they achieved ORR or not (11.73 months vs. 10

| DISCUSSION
In this study, we present a large retrospective study of 156 patients with CMML in China. Our results suggest that HMAs therapy with or without chemotherapy has a superior outcome. Median age at diagnosis in our cohort was younger than the published literature. [2][3][4][5] Almost half of the patients had splenomegaly and nearly one-third of the patients had lymphadenectasis. The LDH (317 U/L) was much higher than normal. The majority of the patients had one or two dysplasia lineages in bone marrow. From the proportion of patients, MP-CMML patients accounted for the majority.
Similar to previous studies, [12][13][14]24 two-thirds of our patients had normal karyotype, while one-third of the patients had chromosome abnormalities in our study. The most frequent cytogenetic abnormality was complex karyotype, so quite a few of patients were categorized into the high-risk group.
From the treatment history, most patients diagnosed before 2010 received chemotherapy. After that, decitabine was widely used to treat CMML alone or combined with chemotherapy in our hospital. Since 2017, azacitidine was applied to the treatment of CMML due to its entry into medical insurance in China. As a result, the majority of patients received decitabine treatment and only a few patients received azacytidine treatment in our study.
In recent years, several trials have accessed the efficacy of HMAs treatment in CMML patients. ORR ranged from 25% to 70% and CR ranged from 10% to 58% in these studies. 21 Median OS ranged from 12 to 37 months. These studies indicated a definite benefit of HMAs in CMML. 25 In our study, the ORR was 79.5% and CR was 8.1% in the HMAs monotherapy group. The ORR was 65.9% and CR was 12.2% in the HMAs ± chemotherapy group. The patients treated with HMAs ± chemotherapy and HMA monotherapy achieved significantly higher ORR than the chemotherapy group. Taken together, HMA-based treatment demonstrated an obvious advantage in comparison to chemotherapy in our study.
Our study showed that the HMAs monotherapy group had prolonged OS than the chemotherapy group. Conversely, the OS and LFS of patients in the chemotherapy group had no relationship with whether they achieved ORR or not probably because they all have poor prognosis.
By univariate analysis, only higher hemoglobulin (≥80 g/L) and lower serum LDH levels (<300 U/L) predicted for better OS and LFS. By multivariate analysis, only Hb ≥ 80 g/L retained prognostic significance on OS, Hb ≥ 80 g/L, and monocytes < 3 × 10 9 /L retained prognostic significance on LFS. These results were like other previously published studies. 25,26 Despite allo-SCT is the solely potentially curative therapy, only 17 patients (10.9%) went through allo-SCT after HMA treatment or chemotherapy in our cohort. The patients Our study has some limitations. First, our data represented a single-center retrospective study. Moreover, because we incorporated into the study with CMML patients diagnosed from 2003 to 2019, only 13 patients had gene sequencing data (NGS available since 2015). Therefore, we could not estimate the prognosis of all our patients with CPSS-mol stratification which integrates conventional indicators and gene mutations, ameliorates the risk stratification of CMML. 7 Finally, we could not assess response with the 2015 International Consortium Response Criteria for Myelodysplastic/Myeloproliferative neoplasm 27 which contains "clinical benefit" because detailed information in accordance with the criteria, were unavailable in most patients.
In conclusion, our retrospective study showed an advantage of HMAs therapy in CMML with ORR of 65.9%-79.5% compared with chemotherapy. The patients who respond to HMAs treatment had longer survival compared to those who did not respond to HMAs treatment. HMAs monotherapy group had prolonged OS than the chemotherapy group. Although HMAs treatment achieved high response but could not significantly modify the disease process. There is still an unslaked need for other therapy that could improve response rates, alter the disease process and prolong survival.

AUTHOR CONTRIBUTION STATEMENT
Liya Ma designed the study and wrote the paper; Lingxu Jiang, Wenli Yang, and Yingwan Luo collected the data; Chen Mei, Xinping Zhou analyzed the data; Gaixiang Xu, Weilai Xu, and Li Ye did the statistical analysis; Yanlin Ren, Chenxi Lu, and Peipei Lin completed the follow-up; Jie Jin edited the paper; Hongyan Tong reviewed and revised the paper.