Long Term Discontinuation of Eltrombopag after Remission in Primary Immune Thrombocytopenia: 8-Year Follow-up Data from 15 Spanish Centers

Background: Successful discontinuation of eltrombopag in certain immune thrombocytopenia (ITP) patients after complete response has already been demonstrated. However, the frequency of this phenomenon and type of candidate patients are still matter of discussion. Moreover, possibility of long term discontinuation responses is not clearly established.

Methods: Here we retrospectively evaluated our whole cohort of 508 adult patients (aged 18 years or more) with primary ITP treated with eltrombopag included in the Spanish Eltrombopag Registry with a focus on the patients who achieved a durable (at least six months) platelet response after stopping eltrombopag. Successful discontinuation of eltrombopag (SDOE) was defined as those patients who reached remission and maintained platelet counts ≥ 50x10⁹/l for at least 6 months in absence of eltrombopag or any rescue therapies administered. Long term discontinuation of eltrombopag (LTDOE) was defined as those patients who reached remission and maintained platelet counts ≥ 50x10⁹/l for at least 36 months in the absence of eltrombopag or any rescue therapies administered. The study was approved by the Hospital Universitario de Burgos Ethics Committee and fulfilled Helsinki declaration standards.

Results: While 37.4% of our patients relapsed of ITP with subsequent platelet count drop sometime during first six months of discontinuation of eltrombopag, a total of 74 patients (14.6%) were able to achieve SDOE.

The median age of SDOE patients was 62 [range, 47-79] years. There were 47 women and 27 men. According to the standard definition, patients were allocated to newly diagnosed (n=17), persistent (n=15) and chronic (n=42) ITP groups. The median time from diagnosis to eltrombopag initiation was 31 [range, 4-104] months. The median number of previous therapies was 2 [range, 1-2], including splenectomy (14%), rituximab (18%) and romiplostim (12%). As expected, all patients but 1 achieved a complete response (platelet count ≥100 x 10⁹/L) prior to eltrombopag discontinuation The median duration of eltrombopag treatment was 7 [range, 2-19] months. Reasons for eltrombopag discontinuation were: persistent response despite a reduction in dose over time (n=43), platelet count >400x10⁹/L (n=16), aspartate aminotransferase elevation (n=5), diarrhea (n=4), thrombosis (n=3), patient's request (n=2) and other reasons (n=1).

Analysis of these SDOE discontinued patients show that with a median follow-up of 55 [range, 29-79] months, 38 patients (51.3%) maintained treatment-free response 36 months after stopping eltrombopag with no need of additional ITP therapies (median time of eltrombopag discontinuation was 70 [range, 50-77] months).This condition is what we define now as LTDOE. Nevertheless, 36 patients relapsed beyond 6 months but before 36 months of eltrombopag discontinuation (median time of eltrombopag discontinuation was 10 [range,7 -22] months).

Characteristics of LTDOE population were a median time since ITP diagnosis of 32 [range, 5-88] months with 15/38 patients having ITP <1 year. 9 patients (24%) were male and their median age was 50 [range, 37-64] years. They had received a median of only two previous treatment lines [range: 1-2 lines]. The median platelet count before starting eltrombopag was 19 x 10⁹/L [range, 8-40]. Meanwhile, platelet count before eltrombopag stop was 218 x 10⁹/L [range, 123-356].

The main characteristics (age, gender, duration of ITP, prior ITP lines, platelet count before starting eltrombopag, duration of eltrombopag treatment, and platelet count before eltrombopag withdrawal) of the 38 patients with LTDOE were compared with those of the SDOE cohort who did not achieve a LTDOE. Unfortunately, no predictive factors of LTDOE could be identified.

Conclusion: Durable platelet response following eltrombopag cessation may be observed in only 15% of primary ITP patients treated with this drug. On the contrary, half of patients who achieve a sustained response after eltrombopag withdrawal will get a long term discontinuation. However, we are lacking predictor factors for successful and long-term discontinuation of eltrombopag in primary ITP.