Venetoclax–proteasome inhibitor–dexamethasone for unfit patients with blastic plasmacytoid dendritic cell neoplasm

TO THE EDITOR:

Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare aggressive hematologic malignancy carrying a very poor prognosis. Intensive chemotherapies using acute myeloid leukemia (AML)– or acute lymphoblastic leukemia–like regimens produce the best response rates but are not feasible in older or unfit patients.¹ Irrespective of the treatment strategy, relapses are the rule, and allogeneic stem cell transplantation is currently considered the only potentially curative option in BPDCN.² In the older or unfit population, who represent the vast majority of cases, therapeutic options are very limited, and survival usually does not exceed few months.³ Hypomethylating agents have been tried alone or in combination with venetoclax. Their clinical efficacy remains modest, is often short-lived, and treatment is often associated with hematologic toxicity that may result in severe infectious complications and death.^4,5 An alternative to chemotherapy is tagraxofusp, an immunotoxin targeting CD123⁺ BPDCN blasts, which demonstrated a 90% overall response rate in the first line and was approved by the US Food and Drug Administration in 2018 for the treatment of treatment-naïve and previously treated patients.^6,7 However, tagraxofusp is not available in many countries and is not exempt from toxicity, notably capillary leak syndrome, which can be arduous to manage in frail individuals.^8,9 

Both normal and malignant plasmacytoid dendritic cells are sensitive to bcl-2 antagonists and proteasome inhibitors.^10-13 Although limited in number, several clinical observations have corroborated these experimental findings either with venetoclax as a single agent^10,14,15 or with hyper-CVAD¹⁶; and a triplet combination with tagraxofusp and azacytidine is currently being evaluated in AML and BPDCN. Similarly, bortezomib in association with lenalidomide and dexamethasone was associated with clinical responses.^17-19 

Based on these results, we hypothesized that combining venetoclax with proteasome inhibitors would increase the efficacy of either agent in BPDCN while limiting the hematologic toxicity and infectious complications that may be frequently observed with the azacytidine and venetoclax combination. Here, we report the results obtained in a retrospective series of 12 frail patients with BPDCN treated with this regimen.

We retrospectively collected data issued from 9 French institutions that treated newly diagnosed patients with BPDCN with a combination of venetoclax, bortezomib, and dexamethasone (dexamethasone was omitted in 1 patient with a psychiatric history) as frontline between June 2022 and February 2024. Patients received continuous venetoclax (400 mg daily orally) combined with subcutaneous bortezomib 1.3 mg/m² on days 1, 4, 8, and 11 of 28-day cycles and dexamethasone 20 mg weekly on bortezomib administration days. A systematic lumbar puncture was proposed for all patients, with a prophylactic intrathecal injection of methotrexate and cytarabine at cycle 1, which was repeated weekly until complete blast clearance if cerebrospinal fluid was infiltrated. Treatment was continued until progression, except in 1 patient who stopped treatment because of concomitant cardiac and infectious complications. Relapse-free survival and overall survival were estimated using the Kaplan-Meier method. Median follow-up time was calculated using the reverse Kaplan-Meier method. All median survival times were provided with a 2-sided 95% confidence interval (Brookmeyer formula). Statistical analyses were performed using SAS software, version 9.4 (SAS Institute, Cary, NC).

All patients gave an informed consent according to local institutional policies.

BPDCN diagnosis was made in 12 consecutive patients after either a skin biopsy or bone marrow aspirate. Our cohort was representative of the epidemiology of BPDCN, with a median age of 73.6 years (range, 49-84) and a strong male predominance (91.7%). At diagnosis, 83.3 % of patients had cutaneous localizations, 41.7% had central nervous system infiltration, and all patients except 1 had bone marrow infiltration. A baseline positron emission tomography scan was performed in 7 patients, which showed a nodal hypermetabolism in 4. Molecular analysis was available in 9 cases. Consistently with previously published data, the most frequent mutations were TET2 (77.8%), ASXL1 (44.4%), and SRSF2 (44.4%). Two TP53 mutations were detected in 1 patient. One patient had no detectable mutations. Table 1 summarizes the patients’ characteristics.

The overall response rate was 100%. All patients reached a complete cytological remission, with complete (CR) or incomplete hematologic recovery (CRi) or clinical CR when presenting initially with isolated skin lesions. All hematologic responses, with a complete disappearance of bone marrow blast infiltration, were obtained after the first cycle of treatment. A cutaneous response was observed within 2 to 4 weeks in 9 of the 10 patients with skin involvement. One patient had incomplete hematologic remission and a partial clinical response, with a residual skin lesion after 1 cycle. Among the 4 patients who had a positron emission tomography scan after C1, all had a complete metabolic response.

Response to treatment and outcome are shown in Figure 1. The median number of cycles was 7 (range, 2-14). With a median follow-up of 14.5 months (95% confidence interval, 2.8-16.7), the relapse-free survival and overall survival were 8.4 and 9.4 months, respectively. Among the 5 patients who presented with central nervous system involvement, 3 were in CR at last follow-up, and 2 had relapsed (with 1 early relapse at 2 months). At the last follow-up, 6 patients were still alive, including 4 in CR. Three patients died due to disease relapse on therapy. One patient died from a posttraumatic brain hemorrhage associated with grade 4 thrombocytopenia while receiving a second-line treatment with 5-azacitidine and venetoclax for a relapse at 3 months. One patient died from a mesenteric infarction at 4 months on therapy while still in CR. One patient stopped the treatment after 4 months while still in CR because of an acute myocardial infarction complicated with cardiac failure and severe infections. He eventually relapsed off-therapy and died 3 months later.

Treatment was well tolerated in this frail population. After initial hospitalization for supportive care when required, it could be administered in an outpatient setting. Treatment-related grade 3/4 cytopenias were observed in only 2 patients (16.7%). Consequently, there were fewer severe infections requiring hospitalization than what was reported with the 5-azacitidine and venetoclax combination in AML.²⁰ Cardiovascular events occurred in 3 patients (27%): 1 case of well-controlled hypertension, 1 mesenteric infarction leading to death, and 1 grade 3 non-ST elevation myocardial infarction requiring hospitalization in an acute cardiac unit. Although all 3 patients had a past history of cardiovascular disease and hypertension, we cannot rule out that proteasome inhibitor may have contributed to these complications. In 1 patient, a grade 2 peripheral neuropathy led to the interruption of bortezomib. The patient continued venetoclax monotherapy and remains in CR 16 months after the diagnosis.

BPDCN still represents a very hard to treat hematologic malignancy, especially in the older population whose life expectancy is ∼4.5 months after diagnosis.³ Our treatment strategy of combining venetoclax, bortezomib, and dexamethasone was associated with rapid and deep responses in patients considered ineligible for intensive chemotherapy. In this frail population, the treatment was well tolerated, in particular without the frequent hematologic toxicity and infectious complications that are observed with venetoclax-azacytidine combination or other cytotoxic agents, and could be administered in an outpatient setting. We believe this chemotherapy-free approach may represent an interesting therapeutic alternative in BPDCN when tagraxofusp or other anti-CD123 targeting therapies are not available or feasible⁹ and might even be considered as a bridge to reduced-intensity conditioning allogeneic stem cell transplantation in selected patients.

Contribution: D.G. and E.D. designed and supervised the study, collected, and analyzed the data and wrote the manuscript; S.K.-H. and A.P. collected and analyzed the data and wrote the manuscript; S.B., R.B., E.A., C.B., A.C., A.D., S.d.B., M.d.C., P.Y.D., A.J., A.M.-K., and C.W. collected the data; and all authors read, revised and proofread the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: David Ghez, Service d’Hématologie, Gustave Roussy, 114 rue Edouard Vaillant, 94805 Villejuif, France; email: david.ghez@gustaveroussy.fr; and Eric Deconinck, Service d’Hématologie, CHU Besançon, 25030 Besançon Cedex, France; email: edeconinck@chu-besancon.fr.