Evaluating Complete Remission with Incomplete Hematologic Recovery (CRh) As a Response Criterion in Myelodysplastic Syndromes (MDS)

Introduction: Myelodysplastic syndromes (MDS) are characterized by ineffective malignant hematopoiesis; current treatments may provide symptomatic improvement and prolong life but are not curative without allogeneic transplant. Hypomethylating agents (HMAs) may prolong survival but have heterogeneous bone marrow blast and hematologic responses. Prior studies have shown the IWG response to HMA is associated with survival, with patients achieving complete remission (CR) having the best OS. It is not clear to what degree other responses may also be associated with survival and therefore may be meaningful outcomes with prognostic significance. The 2006 IWG response criteria include CR, partial remission (PR), marrow complete remission (mCR), stable disease (SD), and progressive disease (PD), as well as hematologic improvement (HI). Recently, CR with incomplete hematologic improvement (CRh, blasts < 5%, ANC > 500, Plt > 50) has been evaluated as a response in acute leukemias, but its relevance is unknown in MDS. We therefore sought to assess what MDS clinical characteristics and patient outcomes are associated with a CRh response following HMA.

Methods: We retrospectively identified adult patients (18+) with MDS (<20% blasts) treated with either azacitidine or decitabine at Massachusetts General Hospital and Moffitt Cancer Center. We assessed best overall response to HMA therapy according to IWG 2006 criteria and noted marrow response (CR, PR, mCR, SD, PD) and hematologic improvement. We subsequently evaluated patients to determine whether they met CRh criteria (bone marrow blasts <5%, ANC > 500 cells/uL, Plt > 50k/uL), and recoded responses according to modified criteria (CR, PR, CRh, mCR, SD, PD). We assessed patient and disease characteristics associated with CRh. We evaluated the duration of HMA therapy according to modified response criteria, from the time of HMA start until treatment cessation. We also assessed overall survival according to response, starting at the time of treatment start, and until death or censored at last known alive using the method of Kaplan and Meier.

Results: We identified a total of 311 patients with MDS who received treatment between 2007 and 2018. The median age at the time of HMA therapy was 69 years (range 23-91). The median ANC was 1.3 (range 0 - 22.3), hemoglobin was 9.5 (4.0-14.0), and platelets were 76 (7-868). Median bone marrow blasts were 5% (0-19%). Most patients had excess blasts (n=181; 58%) and 71% had intermediate or higher risk MDS by IPSS-R (Table 1). Most received azacitidine chemotherapy (n=238, 77%). Median follow up was 60 months.

According to IWG 2006, responses to HMA included CR (n=43, 14%), PR (n=2, 1%), mCR (n=57, 18%), SD (n=149, 48%) and PD (n=60, 19%). A total of 79 patients (25% of total) achieved some form of HI. When adding CRh as a response, a total of 62 patients (20%) met CRh criteria, with concordant decreases in the number classified as mCR (n=26, 8%), and SD (n=118, 38%).

The duration of time on HMA therapy was assessed according to modified response. Patients achieving CR had the longest duration of time on HMA therapy (median 8.1mo) compared to CRh (6mo, HR 1.4, 95%CI 0.9-2.0), mCR (4.7mo, HR 1.9 [1.2-3.1]), SD (4mo, HR 2.0 [1.4-2.9]), PR (4.1mo, HR 3.3 [0.8-13.8]) or PD (2.9mo, HR 5.0 [3.3-7.7]) (p<0.001).

Overall survival confirmed prior studies showing patients with CR had improved OS compared to other IWG 2006 response outcomes (Figure 1A, median 23.3mo, p<0.001). Including CRh as an outcome, OS again varied according to response; median OS was similar between CR (23.3mo) and CRh (25mo, HR 1.0 [0.6-1.6]) (Figure 1B), which was longer than those with responses of mCR (17.2mo, HR 1.3 [0.7-2.2]), SD (16.3mo, HR 1.3 [0.8-1.9]), and PD (8.7mo, HR 2.7 [1.7-4.2]) (p<0.001). Other variables associated with OS on univariate analysis included IPSS-R (p<0.001), while WHO subgroup (p=0.12) and HMA choice (p=0.65) were not associated with survival.

Discussion: Although CR has been the most robust clinical response metric for prediction of OS, MDS patients who achieve a CRh equivalent response had similar survival to those with CR and improved duration on therapy compared to other responses. These data support further evaluation of CRh or other combined marrow/blood count endpoints into future response criteria iterations. These data also suggest new endpoints which could be evaluated in ongoing prospective studies, such as those evaluating novel HMA doublets.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal