Switching ITP patients to avatrombopag from other TPO-ras: A real-world experience of the spanish ITP group (GEPTI).

Abstract Avatrombopag (AVA) is the latest thrombopoietin receptor agonist (TPO-RA). The oral route of administration and the lack of drug and food interaction have made AVA an attractive tool to use instead of other TPO-RAs as second-line therapy to treat immune thrombocytopenia (ITP). A phase 4 study carried out in a 60 ITP patient cohort showed that switching to AVA from another TPO-RA for reasons of convenience was safe and not associated with a drop in platelet count (PC) below 50x109/L. Nevertheless, these observations have not been validated by real-world studies. Nor has the ability of AVA to restore PCs in ITP patients who did not respond to eltrombopag or romiplostim been addressed. We studied the safety and efficacy of switching from eltrombopag and romiplostim to AVA in a cohort of real-world ITP patients recruited by the Spanish ITP Group (GEPTI). Two-hundred and twenty-five adult ITP patients were enrolled in this nationwide, retrospective study. Switching was due to convenience, the occurrence of adverse events (AEs) or lack of effectiveness of the previous TPO-RA in 69, 31 and 123 patients, respectively. The reason was not specified in 2 patients. Switching was from eltrombopag and romiplostim in 131 and 94 cases, respectively. The initial AVA dose was 140 (140, 140) mg/week, median (interquartile range [IQR]). In the overall cohort, 209/221 (94.6%) and 181/221 (81.9%) patients achieved PCs ≥50x109/L (response [R]) and ≥100x109/L (complete response [CR]), respectively. Response could not be assessed in 4 cases because of early AVA suspension (three cases) or early end of follow-up (one case). Considering those patients where switching was due to either convenience or AE occurrence, i.e., those who had responded to the previous TPO-RA treatment (n=100), PCs ≥50x109/L were reported in 97/98 (99.0%) cases and CR was achieved in 91/98 (92.9%) cases (response not assessed in 2 cases). In this subcohort, the follow-up period since AVA was started lasted 446 (222, 718) days. At follow-up end, 90/99 (90.9%) patients persisted on AVA therapy (one missing data). At last control visit, only one patient presented with PCs <50x109/L, and 83/89 (93.3%) showed CR (response not assessed in one patient). AVA dose at last control visit was 80 (60, 140) mg/week. Interestingly, dose was £60 mg in 37/90 (41.1%) cases. Conversely, pre-switch doses of eltrombopag and romiplostim were not below those usually used for maintenance purposes: 350 (175, 425) mg/week and 300 (150, 525) mg/week, respectively. In the subcohort of patients where switching to AVA was due to lack of effectiveness of the previous TPO-RA (n=123), CR was achieved in 88/121 (72.7%) cases, and only 11/121 (9.1%) patients were unable to reach PCs ≥50x109/L (response not assessed in 2 patients). In those who responded, the time required to achieve PCs ≥50x109/L and ≥100x109/L was 14 (6, 32) days and 30 (13, 63) days, respectively. After a follow-up of 427 (156, 673) days since the first AVA dose was administered, 88 out of 110 (80.0%) patients who had responded to AVA maintained PCs ≥50x109/L at last control visit, with CR reported in 58/88 cases. At this time, 94/122 (77.0%) patients persisted on treatment (one missing data), 21 (22.6%), 23 (24.7%) and 49 (52.7%) of whom showed no response, R and CR, respectively (one missing data). The median (IQR) AVA maintenance dose in those who showed R or CR at last control visit was 140 (60, 200) mg/week. The weekly dose was £60 mg in 24/72 (33.3%) cases. Twenty-eight out of 225 (12.4%) patients had treatment-emergent AEs (TEAEs) that resolved with no need of permanent AVA discontinuation. Cephalea, thrombocytopenia in the context of tapering or infection and thrombocytosis were the most frequent ones. In 20/225 (8.9%) cases, TEAEs led to permanent AVA suspension. Venous or arterial ischemic events were reported in 4 cases. Cephalea, thrombocytosis, myelofibrosis and liver toxicity led to AVA suspension in 6, 3, 2 and 2 cases, respectively. Six treatment-unrelated deaths were documented during the study. Collectively, these results support the notion that AVA is a safe and effective tool to use instead of other TPO-RAs to treat ITP, regardless of whether switching is due to patient preference, poor tolerability or inability to restore PCs. The successful use of low doses to maintain response in the long term invites us to consider AVA as a primary option when TPO-RA-based treatment of ITP is scheduled.