Navitoclax Activity Confirmed in Ruxolitinib-Resistant Myelofibrosis

Jacqueline S. Garcia, MD

The addition of the BCL-2/BCL-XL inhibitor navitoclax to ruxolitinib (Jakafi) led to a clinically meaningful reduction in spleen volume and improvement in symptoms in patients with primary or secondary myelofibrosis with acquired resistance to ruxolitinib, according to results of a phase II study presented at the 2019 ASH Annual Meeting.

“This is the first time navitoclax has been added to ruxolitinib,” said lead study author, Jacqueline S. Garcia, MD. “This study serves as proof-of-concept that we may be able to resensitize patients to [JAK inhibition].”

To be eligible for enrollment, patients had to have received ≥12 weeks of continuous ruxolitinib prior to their first dose of navitoclax. In the trial, 34 patients received a daily starting dose of 50 mg of navitoclax and ≥10 mg of ruxolitinib twice daily. Patients were escalated to 300 mg of navitoclax based on tolerability and platelet counts.

Splenic volume reduction from baseline served as the primary endpoint of the trial. Key secondary endpoints included the effect on total symptom score (TSS), overall response rate, rate of anemia response, improvement in bone marrow fibrosis, and safety.

Notably, 52% of patients with available baseline testing had high molecular risk. The prior duration of ruxolitinib therapy was 21 months, and the baseline spleen volume was 1665 cm³. At week 24, 30% of patients achieved spleen volume response of ≥35% (SVR35), and palpable splenomegaly had resolved in 53% of patients. Additionally, 65% of patients experienced alleviated symptoms, and 35% of patients reported a ≥50% reduction in TSS. Moreover, 25% of patients had a reduction in bone marrow fibrosis compared with baseline.

The median duration of navitoclax exposure was 330 days, suggesting that the combination was well tolerated, Garcia explained. Although early-onset grade ≥3 thrombocytopenia was observed in 44% of patients, it was managed with a weekly dose-escalation approach.

In an interview with OncLive, Garcia, an instructor of medicine in the Department of Medical Oncology at Harvard Medical School, and a physician at Dana-Farber Cancer Institute, discussed the design and findings of the phase II navitoclax trial and detailed next steps for the BCL-XL/BCL-2 inhibitor.

OncLive: What was the rationale for the study?

Garcia: Currently, no drugs are approved for use in myelofibrosis beyond the frontline setting. In the frontline setting, we have ruxolitinib and fedratinib (Inrebic), which are both JAK1/2 inhibitors. However, we don’t have any available drugs for patients who develop resistance to these agents. The natural history of the disease is progression. We're looking for drugs that can overcome resistance, and there’s strong preclinical rationale for using the BCL-XL/BCL-2 inhibitor navitoclax.

What differentiates navitoclax from other investigational agents in this space?

Navitoclax is an oral BH3 mimetic that targets cells that are dependent on BCL-XL—less so BCL-2 and BCL-W. What is really nice about this therapy is that we have been able to see that myelofibrosis cells are apoptotically primed, and there’s evidence for the activity of BCL-XL inhibition in reversing fibrosis and causing apoptosis. In preclinical models, we’ve also seen activity with BCL-XL inhibition in JAK2-mutant and -resistant [models]. We have evidence that there’s synergy when BCL-XL inhibitors and JAK2 inhibitors, such as ruxolitinib, are combined. The preclinical rationale is there, and the fact that navitoclax is an oral therapy makes the combination very interesting.

Could you discuss to the design of the trial?

This was a phase II clinical trial that tested the addition of navitoclax to the treatment of patients on standing ruxolitinib. Patients had to have had prolonged ruxolitinib exposure for a minimum of 12 weeks. The median prior time on ruxolitinib was approximately 21 months, which suggests that these were heavily pretreated patients. Patients who came on study had to be on a stable dose of ruxolitinib of at least 10 mg twice daily for 8 weeks, so we knew what their adverse events were from ruxolitinib. We wanted to ensure that patients achieved their best response on ruxolitinib before coming onto study. When navitoclax was added, we assessed patients on a weekly basis. Patients received daily navitoclax at a dose of 50 mg and slowly increased to 300 mg daily if they could tolerate it.

The primary endpoint of the trial was SVR35 at week 24 compared with baseline. Secondary endpoints included a benefit or reduction in TSS and bone marrow fibrosis reduction.

What were the results?

The findings from this study are pretty impressive. Of the 34 patients who were treated on study, we were able to demonstrate the regimen’s tolerability. We saw evidence of thrombocytopenia, fatigue, and diarrhea. We got a sense of what cytopenias might occur. Leukocytosis was able to improve. Hemoglobin generally remained stable across the board for several weeks, even up to week 72. Platelet levels did drop a bit between week 6 and 8. Overall, the mean platelet level between that time where the platelet levels plateaued was about 90,000, which is completely acceptable. Most of these toxicities were low grade. Only 1 isolated incidence of grade 4 thrombocytopenia was reported, but it was reversible with a temporary dose hold.

Patients were able to stay on the drug for a median of 330 days, suggesting that this is not a drug that patients come off of easily. Most patients were able to stay on the drug and dose escalate to 300 mg daily. Responses were seen at all dose levels of 100 mg, 200 mg, and 300 mg, which is wonderful. Based on protocol guidance and platelet levels, patients were able to continue on the combination for the most part.

In terms of clinical benefit, the best SVR among evaluable patients was 43%; that includes the majority of patients with high-risk molecular mutations. Among the cohort of 34 patients, 33 had mutations prior to starting the study. Seventeen of the 33 patients carried at least 1 high-risk molecular mutation. This was a pretty resistant and high-risk population and the majority of those patients were able to achieve SVR35, which was very exciting for us.

Half of the patients with high-risk molecular mutations had 2 or more mutations. There are really good data to suggest that the more mutations a patient has, the more resistant they will be to therapy. Not only did we see SVR in 43% of patients, we noted that responses could have happened beyond 6 months. This suggests that with longer follow-up, we might see [an improvement in SVR]. With assessments and paired questionnaires, we also saw that the TSS was reduced in 65% of patients from week 24 compared with baseline.

Overall, the combination appears to be very exciting. We look forward to more data to confirm the safety of the combination and further evaluate its efficacy. Ongoing efforts are being made to further characterize these patients and to examine [navitoclax] in combination and as a single agent in patients with advanced disease.

Was there anything about the results that surprised you or was unexpected?

One of the things that we learned about navitoclax, based on some of the early clinical trials, is that navitoclax can induce thrombocytopenia. Megakaryocytes are dependent on BCL-XL. We know that older platelets clear. However, we’ve learned that if we do a slow ramp up and recommend a dose based on a platelet level upon entry, we're able to mitigate that thrombocytopenia or that big drop. That has been really effective and important for us.

Many of our patients with advanced myelofibrosis don't have normal platelet levels. Although there was some evidence of thrombocytopenia, it was manageable. There was no evidence of clinically significant bleeding. We were able to manage that event by decreasing the dose of navitoclax, or bringing ruxolitinib down to 10 mg twice daily if needed, which we know is a biologically active dose as a single agent.

Ongoing correlative studies still need to be done based on sample availability. One of the things that we've come to appreciate on study is that patients who remained on the combination by weeks 6 to 8 experienced thrombocytopenia, which is reminiscent of when you start someone on ruxolitinib.

One of the questions that still needs to be answered is whether the addition of navitoclax resensitizes patients to JAK2 inhibition. Many patients benefit from JAK2 inhibition. However, at some point, patients develop resistance as a result of upregulation of that pathway through compensatory mechanisms or acquiring new mutations. It was reassuring that the platelet levels mirrored so much of what we would have seen with ruxolitinib alone, suggesting that something is really happening in that pathway. Exactly what is happening remains to be seen and needs to be defined further.

Harrison CN, Garcia JS, Mesa RA, et al. Results from a phase 2 study of navitoclax in combination with ruxolitinib in patients with primary or secondary myelofibrosis. Blood. 2019;134(suppl 1; abstr 671). doi: 10.1182/blood-2019-130158.