Photo courtesy of ASH
ORLANDO, FL—Investigators are pursuing an upfront approval for eltrombopag in combination with immunosuppressive therapy for the treatment of severe aplastic anemia (SAA).
Based on eltrombopag’s single-agent activity in refractory SAA, they hypothesized that its addition to standard immunosuppressive therapy of horse antithymocyte globulin (hATG) and cyclosporine (CsA) in the first-line setting could improve patient outcome.
And, in a phase 2 trial, it did.
“The addition of eltrombopag resulted in over 20% higher overall response rates and complete response rates for both 3 and 6 months,” said Danielle
Townsley, MD, who presented the data at the 2015 ASH Annual Meeting.
Dr Townsley, of the National Heart, Lung, and Blood Institute, National Institutes of Health, in Bethesda, Maryland, presented the findings as abstract LBA-2.*
The US Food and Drug Administration approved eltrombopag to treat refractory SAA in November 2014, and the European Commission approved it in 2015.
Investigators believed eltrombopag in the upfront, treatment-naïve setting could yield higher overall response rates (ORRs) than the 60% to 70% achieved with standard immunosuppressives worldwide.
“[It was] logical to consider treating patients early at the start of their disease,” Dr Townsley said.
So she and her colleagues conducted an investigator-initiated, phase 2, single-center trial of eltrombopag combined with immunosuppressive agents for first-line treatment of SAA.
Study design and patient population
Patients had to have confirmed treatment-naïve SAA, be a minimum of 2 years old, and weigh more than 12 kg. They were excluded if they had prior immunosuppressive therapy with ATG, alemtuzumab, or cyclophosphamide. They were also excluded if they had liver cirrhosis, AST/ALT more than 5 times normal, or Fanconi anemia.
Primary endpoints of the study were complete response (CR) at 6 months and toxicity. Secondary endpoints included ORR and partial response (PR) rate, survival, clonal evolution, and relapse.
Investigators defined CR as having an absolute neutrophil count (ANC) of 1000/μL or higher, a hemoglobin level of 10 g/dL or higher, and a platelet count of 100,000/μL or higher. They defined PR as blood counts no longer meeting criteria for SAA or CR.
All 92 patients received standard hATG (on days 1 to 4) and CsA (for 6 months). Patients in cohort 1 (n=30) also received eltrombopag at 150 mg daily, starting on day 14 for 6 months.
Patients in cohort 2 (n=31) received eltrombopag at 150 mg daily, starting on day 14 for 3 months. And the 31 patients in cohort 3 started 150 mg of daily eltrombopag simultaneously with the immunosuppressants and continued to receive the drug for 6 months.
Investigators assessed response at 3 and 6 months and planned to follow patients for at least 5 years.
Patients in all cohorts were a median of 32 years (range, 3–82), with 21% being younger than 18. About half were male, 66% had less than 1% of a paroxysmal nocturnal hemoglobinuria clone, 37% had a median neutrophil count less than 200/μL, a median reticulocyte count of 20,000/μL (range, 1600–60,400/μL), and a median platelet count of 9000/μL (range, 0–37,000/μL).
Results
At 3 months, the ORR for the entire population was 81%, and the CR rate was 28%. The ORR was 77% in cohorts 1 and 2 and 92% in cohort 3. The CR rate was 17%, 26%, and 44% in cohorts 1, 2, and 3, respectively.
At 6 months, the ORR for the entire population was 86%, and the CR rate was 37%. The ORR was 80%, 87%, and 95% in cohorts 1, 2, and 3, respectively. And the CR rate was 33%, 26%, and 60%, respectively.
Compared to historic rates for patients on hATG and CsA alone, “the addition of eltrombopag resulted in over 20% higher overall response rates and complete response rates for both 3 and 6 months,” Dr Townsley said.
“And for cohort 3, when eltrombopag is given on day 1, the rate of response in evaluable patients to date appears even higher, with 95% overall response rate at 6 months, of which 60% are complete.”
Dr Townsley also noted that, compared to historical experience, neutrophil recovery was more robust in responding patients treated with eltrombopag. Patients on eltrombopag had a mean ANC of 2253/μL, compared with an ANC of 1716/μL for the historic comparator.
“And likewise, more robust platelet recovery was observed with eltrombopag,” Dr Townsley said, with the eltrombopag-treated patients achieving a mean count of 115,262/μL, compared to a mean of 84,303/μL for the historic group.
She added that, among all eltrombopag-treated patients, the median time to neutrophil recovery was 29 days for an ANC greater than 200/μL and 47 days for an ANC greater than 500/μL. In cohort 3—in which eltrombopag was initiated on day 1—those endpoints were achieved in a median of 8 days and 38 days, respectively.
Patients became transfusion-independent for red cells in a median of 42 days and for platelets in a median of 32 days.
Eltrombopag-treated patients had a 99% overall survival at a median follow-up of 18 months (range 1 – 42) when censored for stem cell transplant. When not censored for transplant, their overall survival was 97%.
Adverse events
“The addition of eltrombopag to ATG and cyclosporine was, overall, well tolerated,” Dr Townsley said. “Few grade 3 to 4 events were attributed to eltrombopag.”
Severe cutaneous reactions in 2 patients caused eltrombopag to be stopped, and 10% of patients had grade 2–3 transaminase and bilirubin elevations.
Bone marrow biopsies revealed no increased fibrosis.
One patient with thymoma died while on study due to encephalopathy. And 2 deaths occurred after hematopoietic stem cell transplant, one with relapsed acute myeloid leukemia and the other from relapsed aplastic anemia.
Clonal evolution occurred in 7 patients, 2 who had achieved CR and evolved in 3 and 30 months. Neither patient had bone marrow dysplasia. One patient’s cytogenetics normalized, and the other had stable disease.
“In our protocol, we define any new cytogenetic abnormality as clonal evolution—we have always done this,” Dr Townsley said.
Of the other 5 patients who evolved, 1 achieved a CR and relapsed, 1 achieved a PR and relapsed, 2 achieved a PR, and 1 had no response. Three of these patients had stem cell transplants, 1 had stable disease, and 1 died of acute myeloid leukemia after stem cell transplant.
The investigators concluded that eltrombopag increases complete and overall hematologic response rates in treatment-naïve SAA patients. Immediate introduction of eltrombopag with immunosuppressant therapy may be optimal, and CR does not appear to prevent clonal evolution.
Investigators are currently in the process of conducting a long-term, serial genomic analysis. The study is open for accrual to an extension cohort.
Eltrombopag is marketed as Promacta in the US and Revolade in most countries outside the US.
Dr Townsley disclosed drug and research funding from GlaxoSmithKline and Novartis, developers of eltrombopag.
*Data in the abstract differ from the presentation.