The Effect of Eltrombopag on Platelet Resistance to Apoptosis: The Role of the Bcl-Xl Pathway

Immune thrombocytopenia (ITP) is typically characterized by increased platelet destruction and reduced platelet production. Eltrombopag and Romiplostim are thrombopoietin receptor (TPO-R) agonists that are known to increase platelet counts in patients with ITP by stimulating thrombopoiesis. Platelets also express TPO-R on their surface, but it is unknown whether the thrombopoietin mimetics (TPO-M) have a direct effect on the circulating platelets. Although controversial, in a very small number of ITP patients, TPO-M agents may increase platelet counts in 2–5 days, earlier than would be expected from de novo megakaryocytopoiesis. Platelet survival is hypothesized to be mediated by two molecular intermediates in an apoptotic pathway, Bcl-xL and Bak. Bcl-xL/Bak protein expression in megakaryocytes is regulated in part by TPO-mediated activation of Akt pathways through Jak2 and Stat5. We hypothesized that an increase in platelet count in the first week of treatment might be mediated by TPO-R signaling, resulting in decreased platelet apoptosis. This study explored whether Eltrombopag or Romiplostim treatment has anti-apoptotic effects on platelets of patients with ITP.

Following a treatment wash out period, 75 mg of Eltrombopag once daily or 10 mcg/kg weekly of Romiplostim was initiated for 2 weeks. Blood counts were measured on days 1, 3, 5, 8, 10, 12, and 15. Platelet function and survival was assessed on days 1, 8, and 15 by: immature platelet fraction (IPF), glycocalicin index, Bcl-xL inhibitor (ABT-737) assay, measurement of Bcl-xL by western blot, measurement of several members of the Bcl-xL Akt mediated, apoptotic pathway by flow cytometry (FACS), bleeding score, measurement of thrombin-anti-thrombin complexes (TATs), and quantification of microparticles.

Eight of 10 patients responded to treatment with Eltrombopag with a platelet count ≥ 50,000/μL, and 6 of the 8 responders at least doubled their counts during the 2 weeks of treatment. All 3 patients treated with Romiplostim responded with platelet count ≥ 50,000/μL. In both treatment groups there was a significant increase in median platelet count (p<0.001), median large platelet count (p<0.01), and median absolute IPF (A-IPF, p<0.01), while there was no significant change in median % IPF. The dose of ABT-737 required to kill half of the platelets in the sample (IC50) in the Eltrombopag group was lower in patients at day 1 than in non-ITP controls, and there was an increase in resistance to apoptosis between days 1 and 8, but these changes did not reach statistical significance. Between days 8 and 15 the IC50 declined to pre-treatment levels. In the Romiplostim group there was no significant difference in IC50 between the control and the patients over the 2 weeks of study. There was no significant correlation between the platelet counts and the IC50 values. FACS analysis of members of the AKT signal transduction pathway revealed increased activation of each of the markers between days 1 and 8, followed by a decrease between days 8 and 15. The levels of Bcl-xL and phosphor-AKT(308) decreased from day 1 to day 15. The other lab tests are pending.

Because the A-IPF increased by less than the platelet increase and because the lifespan of the A-IPF is not known, it is unclear if the platelet count increase is solely a result of increased platelet production. Platelet lifespan may be enhanced by Eltrombopag treatment as there was a parallel albeit transient increase in AKT activation markers and platelet apoptosis resistance in the Eltrombopag group. Treatment with Romiplostim did not appear to affect apoptosis resistance although it did result in transient AKT activation. Our data suggest that platelets are more resistant to apoptosis when the levels of anti-apoptotic factors (eg. PTEN, Phospho-GSK3β) involved in the AKT/Bcl-xL pathway are greatest despite a concomitant increase in pro-apoptotic factors (eg. Bak, Bax). Since both the increased AKT activation and apoptotic resistance returned to baseline at day 15, megakaryocytes and platelets already present at the start of treatment may respond differently than those generated de novo in the presence of TPO mimetics.

Fig. 1

View largeDownload slide

Left: Resistance to apoptosis after treatment with Eltrombopag. Right: Protein levels after treatment with Eltrombopag (E) or Romiplostim (R).

Fig. 1

View largeDownload slide

Left: Resistance to apoptosis after treatment with Eltrombopag. Right: Protein levels after treatment with Eltrombopag (E) or Romiplostim (R).

Close modal

Bussel:Portola: Consultancy; Eisai: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; GlaxoSmithKline: Consultancy, Equity Ownership, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Amgen: Equity Ownership, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Cangene: Research Funding; Genzyme: Research Funding; Immunomedics: Research Funding; Ligand: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Shionogi: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Sysmex: Research Funding.