Hanny Al-Samkari, MD

Sujit Sheth, MD

In this commentary, Hanny Al-Samkari, MD, and Sujit Sheth, MD, highlight key studies in the treatment of nonmalignant hematologic disorders presented at ASH 2019 and share their thoughts on the clinical implications of these findings.

Hanny Al-Samkari, MD:
Advances in the Treatment of Autoimmune Hematologic Disorders
I will primarily discuss advances in the treatment of immune thrombocytopenia (ITP). ITP is an autoimmune disorder characterized by impaired platelet production and destruction of platelets, as mediated by platelet auto‑antibodies, complement, and cytotoxic T‑cells. This leads to thrombocytopenia, an increased risk of bleeding, and impaired quality of life for affected patients. Promising early-phase results in the setting of R/R ITP, an area of unmet need, were presented at ASH 2019 for several novel agents. I will highlight 2 here: 1 is a novel BTK inhibitor, PRN1008, which takes advantage of BTK’s known role in modulating immune responses; the other is sutimlimab, an anticomplement C1s monoclonal antibody that inhibits complement activation by platelet surface autoantibodies, a potentially important mechanism by which platelets are destroyed in ITP.⁴⁵ I will also review updated results from a randomized phase III trial of the thrombopoietin receptor agonist (TPO-RA) avatrombopag in patients with chronic ITP. Finally, sutimlimab was also evaluated in the phase III CARDINAL trial for the treatment of cold agglutinin disease (CAD), a rare autoimmune hemolytic anemia marked by chronic activation of the classical complement pathway, which was also presented at ASH 2019.

Phase I/II Trial of Novel BTKi PRN1008 in R/R Primary or Secondary ITP
The BTK inhibitor ibrutinib, which is a standard-of-care regimen in the treatment of CLL, has been shown to be quite effective at inducing remissions in patients with ITP secondary to CLL.⁴⁶ However, ibrutinib inhibits kinases other than BTK, including those critical for platelet function, and as a result carries a risk for bleeding, which is obviously not suitable for the treatment of ITP. Therefore, development of a BTK inhibitor that does not inhibit platelet kinases is a potentially promising therapeutic option for ITP. An ongoing, open-label, dose-finding phase I/II study presented at ASH 2019 evaluated one such novel BTK inhibitor, PRN1008, which avoids the off-target platelet dysfunction associated with other BTK inhibitors used in the treatment of various hematologic malignancies.⁴⁷ Early studies show that in contrast to ibrutinib, it does not alter platelet aggregation in blood from healthy volunteers or ITP patients.⁴⁸

PRN1008 was evaluated at doses ranging from 200 mg daily to 400 mg BID in 31 adult patients with R/R primary or secondary ITP who had previously responded to ≥ 1 previous therapies.⁴⁷ Patients were required to have severe thrombocytopenia with at least 2 platelet counts < 30,000/μL and no other treatment options available. Continuation on stable corticosteroids or TPO-RA therapy was allowed. The primary endpoint was ≥ 2 consecutive platelet counts > 50,000/μL without rescue medication and separated by ≥ 5 days.

Overall, 39% of patients reached the primary endpoint of 2 consecutive platelet counts > 50,000/μL without rescue. When looking only at patients who received the higher doses of 300 or 400 mg BID, 54% achieved the primary endpoint, suggesting that the optimal dose for treating ITP may be on the higher side. Rates of response this high in patients with highly refractory disease are impressive. Furthermore, PRN1008 was well tolerated with primarily mild, grade 1/2 treatment-emergent AEs, including gastrointestinal AEs and fatigue, and no grade 3/4 treatment-emergent AEs, treatment-related bleeding or thrombotic events, or dose-limiting toxicities.

Based on these impressive efficacy and safety results, I think a pivotal randomized phase III study evaluating PRN1008 for the treatment of R/R ITP is the logical next step in confirming the findings of the smaller phase I/II study and getting this agent closer to use in the clinic. This agent could also be useful in earlier settings, where we currently use TPO-RAs and rituximab preferentially. For example, TPO-RAs achieve response rates of approximately 80% in patients with chronic ITP lasting ≥ 1 year, which is a high bar for any new agent to match or surpass. However, a 54% response rate in patients with very refractory disease, as demonstrated here with PRN1008, could conceivably translate into response rates comparable with that achieved with TPO-RA in patients with less refractory disease.⁴⁹ We will have to wait and see what the data indicate.

Phase I Trial of Sutimlimab to Inhibit Classical Complement Pathway in Patients With Multirefractory Chronic ITP
As mentioned, complement activation by platelet surface autoantibodies is thought to be an important mechanism by which platelets are cleared in ITP and, therefore, presents a potentially promising therapeutic target in the treatment of this disease. Sutimlimab, a monoclonal antibody that targets the C1s subunit of C1 and the first protein in the classical pathway of complement activation, was designed to specifically inhibit the classical complement cascade, while leaving the lectin and alternative pathways intact. This may lead to a lower likelihood of infectious complications with this approach.

In an open-label phase I study, sutimlimab was evaluated in 8 adult patients with chronic ITP (disease lasting > 1 year), inadequate response to ≥ 2 previous therapies, and platelet counts ≤ 30 x 10⁹/L.⁵⁰ Patients were treated with IV infusion of sutimlimab (6.5 g if < 75 kg; 7.5 g if ≥ 75 kg) on Days 0 and 7, then biweekly followed by a washout period. After the washout period, responders were retreated to see if they recovered their previous response. The primary endpoints included safety, pharmacodynamic and pharmacokinetic parameters, and durable platelet response (defined as a platelet count ≥ 30 x 10⁹/L and a > 2‑fold increase from baseline on 2 separate occasions > 7 days apart without bleeding).  

Of the 8 patients treated with sutimlimab, 4 (50%) achieved a durable platelet response (as defined above) by Day 14. The platelet response was rapid, with the mean platelet count climbing within 24 hours to 73.9 x 10⁹/L from a mean count of 26.6 x 10⁹/L at baseline. Platelet count improvements were durable over time with continued sutimlimab treatment, even up to > 4 months. During the washout period, thrombocytopenia returned for the 4 responding patients, and then upon retreatment with sutimlimab, durable platelet responses were restored. This agent was generally well tolerated. 

The most impressive result of this study was that platelet response was so rapid, at < 24 hours. For comparison, IV immunoglobulin can sometimes achieve a relatively rapid platelet response within 1-2 days for ITP, whereas corticosteroids can take 2-3 days and often longer. The prospect of potentially having a therapy that can improve platelet counts in hours is extremely attractive and will certainly fill an unmet need in the treatment of ITP, with the caveat that this was a phase I study in only 8 patients and we need to see if this data holds true in a larger study.

Updated Results From a Randomized Phase III Trial of Avatrombopag in Patients With Chronic ITP
In the first-line setting, chronic ITP is treated with corticosteroids or IV immunoglobulin, with TPO-RA therapy being used in the second-line setting or beyond following relapse.⁵¹ Patients with chronic ITP often take TPO-RA therapy for very long periods of time, often years or indefinitely, to keep their disease under control and their platelet count in an acceptable range. FDA-approved TPO-RAs include eltrombopag, romiplostim, and avatrombopag.

Avatrombopag, the newest TPO-RA developed, was approved in 2019 for the treatment of thrombocytopenia in adult patients with chronic ITP and an insufficient response to previous therapy⁵² based on results from a 6-month randomized phase III trial of 49 patients showing that avatrombopag achieved a superior median cumulative duration of platelet response ≥ 50,000/µL (12.4 vs 0 weeks; P < .0001) and greater platelet response rate at Day 8 (65.6% vs 0%; P < .0001) compared with placebo.⁵³ At ASH 2019, long-term follow-up data of the 39 patients who proceeded to the open-label extension of this phase III study were reported, with alternate endpoints of platelet response (≥ 50,000/µL) or complete platelet response (≥ 100,000/µL) at any time.⁵⁴  

During the open-label extension phase, avatrombopag maintained improvements in platelet response in patients with refractory chronic ITP, showing that continued treatment with this agent is effective over long periods of time. Platelet counts ≥ 50,000/µL and ≥ 100,000/µL were observed for 44.1% and 24.1% of patients receiving continued avatrombopag, respectively; these values are very similar to those achieved with avatrombopag during the core study (48.6% and 25.6%, respectively). For placebo, no patients exhibited a platelet response in the core study, but as would be expected, those who crossed over to avatrombopag for the extension did: 41.3% and 18.3% had platelet counts ≥ 50,000/µL and ≥ 100,000/µL, respectively. Notably, a subset of 7 patients with especially durable responses in the core study (they showed platelet responses during 6 of the final 8 weeks of the study) achieved platelet levels ≥ 50,000/µL and ≥ 100,000/µL with avatrombopag at 96.1% and 60.1%, respectively, of the extension phase visits.

In addition to being effective, avatrombopag continued to be quite well tolerated. AEs similar to other TPO-RAs, such as headache, were observed. However, there was no signal for liver injury (ie, increased alanine aminotransferase and bilirubin levels) as was seen with eltrombopag and for which it has a black box warning for hepatotoxicity.⁵² Notably, this study involved significantly fewer patients.

These updates providing longer-term efficacy and safety data for avatrombopag in patients with chronic ITP are important because in the real world we treat patients longer than the 6 months that was allotted for the study. With regard to safety, there can be AEs that occur at such low rates that they are not captured during the study and only show up when patients are treated for long periods of time, so it is great to see confirmation that avatrombopag is indeed safe and that its efficacy does not drop off. These data support the use of avatrombopag for extended periods of time in the treatment of patients with chronic ITP.

Although we do not have head-to-head studies to compare the efficacy of the available TPO-RA therapies, there are some differences among these agents to note. Eltrombopag and avatrombopag are both oral agents and romiplostim is injectable. Eltrombopag requires critical timing restrictions for the types of food that can be taken with it in order to ensure good absorption, including no intake of calcium, magnesium, other minerals, and fat during a 4- to 6-hour window around taking the drug. Avatrombopag does not have these dietary restrictions, which is to its benefit, nor does it appear to be associated with a risk for hepatotoxicity like eltrombopag.

CARDINAL: Phase III Study of Sutimlimab in Patients With CAD and a Recent History of Transfusion
Sutimlimab, the first-in-class inhibitor of the classical complement pathway discussed above, was also evaluated for the treatment of patients with CAD and a recent transfusion history in the open-label single-arm phase III CARDINAL trial.⁵⁵ All 24 enrolled patients received weight-based dosing of IV sutimlimab on Days 0 and 7, then every 2 weeks for 26 weeks. Patients were required to have autoimmune hemolytic anemia, specifically CAD, with a hemoglobin level of ≤ 10 g/dL and have received at least 1 blood transfusion within 6 months of enrollment. The primary endpoint was response defined as an increase in hemoglobin of ≥ 2 g/dL from baseline or achieving a hemoglobin of ≥ 12 g/dL without transfusions or protocol-prohibited therapy for CAD. After the 26-week treatment period, patients could continue sutimlimab therapy every 2 weeks in the safety extension period.

In patients with CAD, sutimlimab achieved a mean hemoglobin increase of 2.6 g/dL over the 26-week treatment period, which is quite impressive in this disease. Furthermore, response coincided with a significant improvement in quality of life as assessed by the FACIT-F scale for fatigue. Again, sutimlimab was well tolerated, with no serious treatment-emergent AEs related to this agent.

The results of CARDINAL poise sutimlimab as a promising new agent for the treatment of CAD, which is a disease currently lacking good standard-of-care treatment options.⁵⁶

Sujit Sheth, MD:

Advances in the Treatment of Hemoglobin Disorders
Treatment options for patients with inherited hemoglobin disorders continue to evolve. At the ASH 2019 annual meeting, we saw promising updates on gene therapy for the treatment of sickle cell disease (SCD), which is caused by a single amino acid substitution in β-globin that causes polymerization of hemoglobin S, and β-thalassemia, which is characterized by impaired β-globin synthesis leading to severe anemia and transfusion dependence. In addition, updates of data from the phase III BELIEVE trial evaluating luspatercept, a new disease-modifying therapy that was recently approved by the FDA⁵⁷ for the treatment of adult patients with β-thalassemia who receive regular blood transfusions, were presented.

Pilot Feasibility Study of BCL11A as a Gene Therapy Target in SCD
The first study we will discuss is a pilot feasibility study of gene therapy for the treatment of SCD, which took the approach of modifying BCL11A expression to increase fetal hemoglobin production.⁵⁸ Specifically, BCL11A represses γ-globin expression postnatally, resulting in a switch from fetal hemoglobin (HbF) to adult hemoglobin (HbA) and enhanced β-globin production. By targeting BCL11A and blocking its function, γ-globin expression is allowed to continue and HbF continues to be produced. This is of benefit to patients with SCD because HbF interferes with polymerization of sickle hemoglobin (HbS) and inhibits cellular sickling, thus ameliorating the severity of the disease. Should this be successful, patients could have mild or no manifestations, depending on the amount of HbF produced. However, it is a stem cell–based treatment, and patients would essentially undergo an ASCT with myeloablative conditioning, which is a procedure not without risks. Thus, the intent of this therapeutic approach must be curative and not simply palliative.

It is planned that 15 patients with severe SCD who did not respond to hydroxyurea and who did not have a matched sibling donor for hematopoietic SCT will be treated in this clinical trial. Peripheral stem cell collection will be performed, from which CD34+ cells will be selected and transduced ex vivo with the BCH-BB694 construct (a shmiRNA lentiviral vector encoding a BCL11A-specific shRNA embedded in a microRNA scaffold that confers erythroid-specific knockdown of BCL11A) to generate the gene therapy drug product. Following myeloablative busulfan conditioning, the drug product will be infused back into patients.

In the 5 patients who have received BCH-BB694–transduced drug product to date, HbF levels ranged from 24% to 43%, with the duration of follow-up ranging from 1-18 months following infusion. Although this will likely ameliorate the clinical severity of the SCD, it may not be curative. Like the other lentiviral gene addition clinical trials (ie, of LentiGlobin BB305–based gene therapy), there will likely need to be a refinement in the process to ensure a greater proportion of stem cells are edited. These data do, however, provide proof of principle that a gene editing approach to target BCL11A and induce HbF expression in patients with SCD would work. Also in the gene editing realm, we await data from a phase I/II trial evaluating the safety and efficacy of the CTX001 drug product (autologous CD34+ stem cells that have been modified at BCL11A using CRISPR/Cas9 technology).⁵⁹

Whereas these trials only have early results, in the long term they will be assessed against other gene therapy approaches that are further along in clinical trials such as LentiGlobin BB305–based gene addition therapy, where a copy of a functional β-globin (β^T87Q) gene is added to the hematopoietic stem cell genome. These trials have shown positive results, with most patients with severe SCD having predominantly functional HbA after treatment.⁶⁰ Whether gene addition or gene editing is a more durable strategy will only become clear in the longer-term as data from these trials mature. 

Interim Results From the Phase III Northstar-3 (HGB-212) Study of Betibeglogene Autotemcel in Patients With Severe Transfusion-Dependent β-Thalassemia
An update on gene therapy for the treatment of β-thalassemia was also presented at ASH 2019. The phase III Northstar‑3 (HGB‑212) study evaluated the LentiGlobin BB305–based gene therapy betibeglogene autotemcel, as produced by a refined manufacturing process, for the treatment of patients with transfusion-dependent β-thalassemia and β⁰β⁰ genotypes, the most severe β-thalassemia genotype.⁶¹ For context, patients with transfusion-dependent β-thalassemia and β⁰β⁰ genotypes in the early-phase HGB-205 study evaluating this agent all experienced a reduction in transfusion burden, but only 3 of 8 patients achieved transfusion independence.⁶² Therefore, to improve clinical outcomes, the drug product manufacturing process was refined for the phase III study.

In Northstar-3, 17 patients aged 6-50 years with transfusion-dependent β-thalassemia and a β⁰/β⁰, β⁰/IVS1-110, or IVS1-110/IVS1-110 genotype, were enrolled, with 16 completing stem cell mobilization and 13 receiving infusion of the drug product. The endpoints for this phase III trial were specific, the primary endpoint being a ≥ 60% reduction in RBC transfusion volume between 12 and 24 months posttransplant. A key secondary endpoint was transfusion independence defined as weighted average hemoglobin ≥ 9 g/dL without transfusions for ≥ 12 months.

The interim results of Northstar-3 showed much improvement over those from HGB-205.⁶¹ All of the 12 patients with ≥ 3 months of follow up were off transfusions at the time of this report; 10 could be considered transfusion dependent since they were < 12 months from treatment. The hemoglobin levels, including hemoglobin with β-globin T87Q expressed from the vector, achieved in Northstar-3 were also improved compared with HGB-205. More mature data, once all patients have been followed for ≥ 12 months, will be critical to assess the success of this approach. It would also be very important to know if the improved results were related to the new approach to processing of the stem cells, or whether there were pretreatment clinical factors predictive for response that would facilitate patient selection in the future.

Betibeglogene autotemcel received its initial approval in Europe for patients with transfusion-dependent β-thalassemia and non-β⁰β⁰ genotypes based in part on the results of the phase III Northstar-2 (HGB-207) study.⁶³ Approval in the United States will depend on the maturity of the data at the time of submission to the FDA. If data from the Northstar-2 and -3 studies hold up, it could be approved for both non-β⁰β⁰ and β⁰β⁰ genotypes. Pediatric patients were included in the later trials and approval would likely be for both adults and children. Younger patients with little to no comorbidity are likely better candidates to undergo myeloablative conditioning and autologous transplant compared with older patients.

BELIEVE Subanalysis: Evaluation of Responders With Transfusion-Dependent β-Thalassemia to Luspatercept
Updates were presented for the randomized phase III BELIEVE trial evaluating luspatercept, a novel erythroid maturation agent, in the treatment of patients with transfusion‑dependent β-thalassemia.⁶⁴ Initial results from BELIEVE presented at ASH 2018 created plenty of excitement because all primary and secondary endpoints were met with statistical significance.⁶⁴ Notably, 70.5% vs 29.5% of patients in the luspatercept arm vs the placebo arm, respectively, had a ≥ 33% reduction in transfusion burden and 40.2% vs 6.3%, respectively, had a ≥ 50% reduction in transfusion burden in any 12-week interval during the study. In addition, 41.1% vs 2.7% of patients had a ≥ 33% reduction in transfusion burden and 16.5% vs 0.9% of patients had a ≥ 50% reduction in transfusion burden in any 24-week interval during the study. This led to the approval of luspatercept by the FDA for the treatment of anemia in adult patients with β-thalassemia who are receiving regular blood transfusions.⁵⁷

The BELIEVE ad hoc analysis reported at ASH 2019 assessed the durability of responses and longer-term safety in patients who continued luspatercept.⁶⁵ Reductions in transfusion burden of ≥ 33% and ≥ 50% over any 12- or 24-week period that were reported in the primary analysis were maintained with longer follow-up. However, only 25 patients (11.2%) achieved transfusion independence for ≥ 8 weeks; 5 patients (2.2%) for ≥ 24 weeks; and 3 patients (1.3%) for ≥ 48 weeks. These results demonstrate that responses were durable among the patients who had the best responses on luspatercept but that, likely, most patients will not achieve transfusion independence with this agent.

The investigators also showed that the safety profile was similar to what was reported at the end of the 48‑week primary analysis. The most common AEs occurring with luspatercept were bone pain (20.2% vs 8.3% with placebo) and arthralgia (21.1% vs 14.7% with placebo), with dizziness being less frequently reported (12.1% vs 4.6% with placebo). However, these were mostly low grade and improved over time.

In terms of clinical applicability of these data, the durability of responses to luspatercept seems quite good. Despite only a small number of patients becoming transfusion independent, the overall reduction in transfusion requirements was quite significant, which will translate into significant improvements in clinical outcomes, in particular a reduction in iron overload and better quality of life, even for patients who do not have the best responses. The implication of having your transfusion requirement go down, even if it does not go down to 0, is that you can receive fewer units of blood each time you get transfused and thereby spend a shorter time at the hospital, or be transfused at a longer interval and miss fewer days of work or school. Furthermore, it is very reassuring that no new safety signals have emerged with longer follow-up.