Title: Durable Remission and Long-Term Survival in Relapsed/Refractory (r/r) AML Patients Treated with Guadecitabine, Median Survival Not Reached for Responders after Long Term Follow up from Phase 2 Study of 103 Patients
Authors: Elizabeth A. Griffiths, MD1, Hagop M. Kantarjian, MD2, Casey L. O’Connell, MD3, Karen W.L. Yee, MD4, Wendy Stock, MD5, Naval G. Daver, MD2, Elias Jabbour, MD2, Ellen K. Ritchie, MD6, Jean-Pierre Issa, MD7*, Katherine J. Walsh, MD8, David A. Rizzieri, MD9, Scott D. Lunin, MD10*, Xiang Yao Su, PhD11*, Mohammad Azab, MD11and Gail J. Roboz, MD6
1Roswell Park Cancer Institute, Buffalo, NY; 2The University of Texas MD Anderson Cancer Center, Houston, TX; 3Keck School of Medicine, University of Southern California, Los Angeles, CA; 4Princess Margaret Cancer Centre, Toronto, ON, CAN; 5University of Chicago Medical Center, Chicago, IL; 6Weill Cornell/NY Presbyterian Medical Center, New York, NY; 7Fels Institute for Cancer Research and Molecular Biology, Temple University, Philadelphia, PA; 8The Ohio State University, Columbus, OH; 9Duke University Medical Center, Durham, NC; 10Sarah Cannon Research Institute, Florida Cancer Specialists, Venice, FL; 11Astex Pharmaceuticals, Inc., Pleasanton, CA
Background: Guadecitabine is a next generation subcutaneous (SC) hypomethylating agent (HMA) resistant to degradation by cytidine deaminase which results in prolonged in vivo exposure to the active metabolite decitabine. We conducted a phase 2 study of guadecitabine in 103 r/r AML patients. We present here duration of response and long-term survival results.
Methods: We conducted a phase 2 study of guadecitabine using different regimens and doses (randomized 5-day regimen cohorts of 60 mg/m2/d vs 90 mg/m2/d SC) and a cohort of 10-day regimen in the first 1-4 cycles at 60 mg/m2/d followed by subsequent cycles of 5-day regimen). Response and duration of response were assessed using IWG 2003 criteria: Complete Response (CR), CR with incomplete platelet recovery (CRp), and CR with incomplete count recovery (CRi). CR+CRp+CRi was defined as composite CR (CRc). Overall survival (OS) was assessed using the Kaplan-Meier (KM) method. Response status for each dose/regimen cohort and the overall treated population were assessed with analyses of duration of response and long-term survival.
Results: The study completed enrolment of 103 r/r AML patients: 50 patients received 5-day regime at 60 mg/m2/d (24 patients) or 90 mg/m2/d (26 patients), and 53 patients received the 10-day (60 mg/m2/d). Median follow up was 2.4 years (29.1 months). Patients’ characteristics for the 103 r/r AML patients enrolled included median age of 60y (range 22-82y), poor risk cytogenetics in 41%, prior hematopoietic cell transplant (HCT) in 18%, median number of prior regimens 2 (range 1-10), primary refractory to induction therapy in 47%, and 41% had a high disease burden of BM blasts >40%. There was no significant difference in CR or OS between 60 and 90 mg/m2/d 5-day regimen but the CR and CRc rates were higher on the 10-day regimen (19% and 30% respectively) vs the 5-day regimen (8% and 16%). When all regimens analyzed together, 24/103 patients (23.3%) achieved CRc. Responses (CRc) were achieved in several poor prognosis subgroups including 19% in patients with poor risk cytogenetics, 31% of refractory patients, 26% of patients who relapsed after prior HCT, and 22% in patients with early relapse (< 6 months from their prior treatment). Of the 24 CRc patients, 15 (63%) were refractory to induction, 8 (33%) had poor risk cytogenetics, and 5 (21%) had prior HCT, and 14 (58%) went on to receive HCT following response. Median overall duration of response for patients with CR, and CRc were 7 and 7.8 months respectively. After long term follow up, median OS has not been reached in patients who achieved CRc (either CR or CRp/CRi). The 2-year survival rate was 57% for CR, and 50% for CRp/CRi (Fig. 1). Median OS has not yet been reached and was similar in CRc patients who went on to receive HCT post CRc (14 patients) compared to CRc patients who did not receive HCT post treatment (10 patients) (Fig.2). The 2-year survival rate was also similar for both groups (50% for those receiving HCT vs 60% for those who did not undergo HCT). Most patients were still on guadecitabine treatment until death, progression, or HCT with no other subsequent treatment. Guadecitabine was well tolerated in all cohorts with Grade 3 or higher AEs related to the drug seen in 42% of patients predominantly myelosuppression and related infections. There was no related serious AEs leading to death. The results highlight the long survival benefit for guadecitabine responders that exceeds duration of response and seems irrespective of post treatment HCT. The results also suggest that in r/r AML patients treated with guadecitabine, CRp/CRi seem to confer a similar survival benefit to CR patients suggesting that the incomplete peripheral blood count recovery may reflect continued treatment-related myelosuppression rather than active residual disease.
Summary/Conclusions: In a phase 2 study of HMA guadecitabine in heavily pretreated r/r AML patients, 47% of whom had refractory disease, CR, CRp, and CRi all conferred long survival benefit. With a median follow up of almost 2.5 years, more than half of responding patients were still alive at 2 years and their median OS has not yet been reached. In addition, treatment with guadecitabine allowed post treatment HCT in 58% of responders.