Title: <scp>Early‐salvage</scp> therapy with venetoclax‐based regimens for induction failure and poor early response acute lymphoblastic leukaemia: A retrospective case series of 13 patients
Abstract: The achievement of early and deep response is an essential step for improving the prognosis in acute lymphoblastic leukaemia (ALL). However, 5%–25% of adult ALL patients show a poor early response or induction-therapy failure.1-3 Among these patients, resistance to the re-use of conventional chemotherapeutic agents is commonly observed.4 Meanwhile, serious complications, heavy tumour burden, and low affordability may reduce the advantage of immunotherapeutic approaches, such as chimaeric antigen receptor-T cell (CAR-T) therapy and bispecific T-cell-engaging antibodies to be optimal early-salvage or induction agents.5-7 Venetoclax, a B-cell lymphoma 2 (BCL2) inhibitor, which achieves encouraging results in acute myeloid leukaemia, represents another opportunity in ALL as preclinical data demonstrate promising activity in xenograft ALL models.8-10 Nevertheless, whether incorporating venetoclax into early-salvage treatment even as an adjuvant agent in the late phase of induction could induce an early and deep response, and improve the long-term outcome of adults ALL patients, is still unknown. Herein, we retrospectively reviewed 13 consecutive cases of adult ALL with induction failure or poor early response who were early salvaged with venetoclax-based regimens. From March 2021 to July 2022, seven adult patients with induction failure (defined as failure to achieve <5% blasts in bone marrow after induction therapy) and six with poor early response (defined as >30% blasts in bone marrow at day 14 of induction therapy) salvaged with venetoclax-based regimens were enrolled in this study. Among them, three patients were diagnosed as T-cell ALL (T-ALL), including two early T-cell precursor (ETP)-ALL; the other 10 were B-cell ALL (B-ALL), including two Philadelphia chromosome-like (Ph-like) ALL. All the patients received scheduled therapy following the trial protocol at our centre (ChiCTR-TNC-09000397). The median age was 38 years old (range, 26–72 years). For the tumour burden, the median white-cell count and bone-marrow blast percentages were 8.5 (range, 1.71–177.2) × 109/l and 56.0% (range, 15%–90%) before treatment, respectively. High-risk chromosome aberrations, gene fusion, mutation, and subtype had been detected in nine (69.2%) patients (Table 1 and Supplementary material). Venetoclax was incorporated into day 15 of induction therapy for patients with poor early response and within re-induction treatment for patients who failed for the first or second rounds of induction therapy. The dose of venetoclax was 200 mg/d for 7 or 14 days, which depended on the combined agents. Only one patient was given venetoclax 100 mg/d for seven days with concomitantly the azole antifungal posaconazole (Table S1). Notably, the addition of venetoclax to the early-salvage regimen was particularly effective in six patients who did not achieve early bone-marrow response on day 14 of induction. For these patients, the median blast percentage at day14 of induction therapy is 76.7%. Encouragingly, all six patients achieved complete remission (CR) with four obtaining minimal residual disease (MRD)-negative CR after one cycle of treatment, including one ETP-ALL and one Ph-like ALL. Three patients subsequently underwent haematopoietic stem cell transplantation (HSCT) and all six patients are alive in remission at the time of writing. This indicates the application of venetoclax in pre-emptive salvage treatment may induce a deeper response in patients with poor early response. This result is consistent with the previous data that combining venetoclax with dose-reduced chemotherapy could induce a high MRD-negative remission rate in the frontline treatment for older ALL, which also supports our strategy of employing venetoclax in the up-front salvage therapy for deeper response.11 Among the seven patients who failed the induction therapy, four (57.1%) patients achieved CR after one cycle of treatment. One experienced relapse and was treated with CAR-T therapy as second salvage, and three cases underwent HSCT subsequently after venetoclax-based therapy (Figure 1A). Although doses of venetoclax were the lowest dose, 200 mg/d and a seven or 14 days medication cycle, the CR rate in our series is higher than in historical intensive chemotherapy (11%–40%).12 Meanwhile, this result also is comparable with refractory ALL patients who received venetoclax 400 mg daily in combination with navitoclax (a BCL2/BCL2L1/BCL2L2 inhibitor), although the number of patients remains limited.13 After a median follow-up of 7.2 months, the median overall survival (OS) and the relapse-free survival have not been reached for the whole cohort (Figure 1B,C). All 10 patients achieving CR remain alive in remission, at last time of follow-up. Two of three patients without response died of disease progression. The median OS was 8.0 months for non-responders and has not yet been reached for responders (Figure 1D, p = 0.009). The OS would remain longer in those patients who have undergone HSCT than with conventional chemotherapy (Figure 1E, p = 0.02). This benefit may relate to venetoclax-induced leukaemia-free status, given a deeper remission state is an extremely important prognostic factor for HSCT.14 The results indicate employing venetoclax-based regimens as early salvage could provide a bridge to subsequent HSCT in induction failure and poor early response in patients with ALL, which may transform into long-term benefits for them. Toxicities were graded according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. No early death or tumour lysis syndrome was observed. The median granulocyte deficiency time, defined as the median number of days during which the neutrophil count is less than 0.5 × 109/l, was eight days (2–13 days) after venetoclax therapy. The median time of platelet counts below 25 × 109/l was 0 days (0–9 days). Only six patients required platelet transfusions. Non-haematological adverse events (AEs) mainly included hepatotoxicity (grade 1), pneumonia (grade 3), and febrile neutropenia (grade 3), each observed in two patients, respectively. There was one case of grade 3 intestinal infections followed by sepsis (Table S2). Of note, all toxicity was reversible and there was no need to reduce the treatment dose or for discontinuation in each cycle, even in combination with regular-dose chemotherapy. Overall, our research firstly demonstrated that venetoclax-based early-salvage therapy could achieve an ideal response with a well-tolerated safety for induction failure and poor early-response ALL. This empirical attempt may lay a foundation for employing venetoclax as a pre-emptive salvage regimen, to induce an early and deep remission and provide a chance for subsequent HSCT, which may further improve long-term outcomes for high-risk patients. The addition of venetoclax may be considered as a new therapeutic option for up-front/frontline salvage therapy, though larger groups of patients are needed to verify this. Yingchang Mi and Jianxiang Wang served as the principal investigators for this study. Shuning Wei performed the research, analysed the data and wrote the paper. Runxia Gu, Yang Song collected and assembled the data. Zhangsong Yan, Yimin Hu, Dong Lin, Kaiqi Liu, Chunlin Zhou, Guangji Zhang and Ying Wang contributed to patient recruitment and treatment. All authors participated in the critical review and revision of this manuscript and provided approval of the manuscript for submission. This research was supported by grants from the National Key Research and Development Program of China (2019YFC0840605, Grant to Yingchang Mi). All authors declare that they have no conflict of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request. Data S1 Table S1 Table S2 Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.