Long-term and low-dose of busulfan is a safe and effective second-line treatment in elderly patients with essential thrombocythemia resistant or intolerant to hydroxyurea

Essential thrombocythemia (ET) is a Philadelphia negative myeloproliferative neoplasm (MPN) characterized by an increased risk of thrombosis and, in the long-term, transformation to myelofibrosis (MF) or acute leukemia. The purpose of therapy is to achieve a good cytoreduction and to prevent thrombosis and bleeding complications without increasing the transformation risk. In patients at high risk of thrombosis, cytoreduction with hydroxyurea (HU) is currently recommended as first-line therapy. Busulfan (BU) is an alkylating agent that has been used since 1959 for the treatment of MPN. In some retrospective studies, its use according to the conventional schedule has been associated to high risk of leukemic transformation and second neoplasms, as with other alkylating agents. Consequently, BU is currently used by many hematologists as second-line treatment in Bcr-Abl negative MPN that are resistant to or developed side effects from HU, principally in elderly patients, for whom therapeutic options remain limited. The conventional schedule provides a starting dose of 14mg/week up to obtain a complete haematological response (CHR), defined according to the European LeukemiaNet (ELN) criteria . In the current retrospective study, we wanted to share our experience with use of long-term and low-dose of BU in elderly patients with ET who are resistant or intolerant to HU. From 1995 to 2015 we collected 348 patients who received ET diagnosis at our department, according to WHO criteria. The principal clinical and hematological features of entire cohort of ET patients were shown in Table 1. In this population we identified 30 patients who received a second-line treatment with an alternative schedule of BU, defined by low-starting dose (4–6mg/ week) down to obtain CHR, followed by a dose deescalation of BU over time up to the minimal dose required for the maintenance of CHR. The dose deescalation of BU was performed every 4 weeks and the minimal dose to maintain the CHR was generally between 2mg/week and 2mg/month. We analyzed efficacy, toxicity and risk of leukemic evolution in these settings of patients. Intolerance and resistance to HU were categorized according to ELN criteria for ET. Non-parametric tests, such as Mann–Whitney, Pearson Chi-square and Fischer’s exact tests, were used for statistical analysis of continuous and categorical variables. Leukemia-free-survival (LFS) curves were calculated by Kaplan–Meier method and compared with Log-rank (Mantel-Cox) test. Twenty-six of 30 patients were evaluable for full demographic, diagnostic and therapeutic information and were considered for statistical analysis. As reported in Table 1, we observed some significant differences between BU patients and the entire series of ET patients with respect to age and blood counts at diagnosis. The BU patients were older at diagnosis and presented higher platelet (PLT) count and lower levels of hematocrit (Ht), as compared to global ET population. These demographic

Long-term and low-dose of busulfan is a safe and effective second-line treatment in elderly patients with essential thrombocythemia resistant or intolerant to hydroxyurea Rossella Renso 1,2 , Andrea Aroldi 1,2 , Pietro Pioltelli 1 , Carlo Gambacorti-Passerini 1,2,3 and Elena Maria Elli 1 Essential thrombocythemia (ET) is a Philadelphia negative myeloproliferative neoplasm (MPN) characterized by an increased risk of thrombosis and, in the long-term, transformation to myelofibrosis (MF) or acute leukemia 1 . The purpose of therapy is to achieve a good cytoreduction and to prevent thrombosis and bleeding complications without increasing the transformation risk. In patients at high risk of thrombosis, cytoreduction with hydroxyurea (HU) is currently recommended as first-line therapy 2 . Busulfan (BU) is an alkylating agent that has been used since 1959 for the treatment of MPN 3 . In some retrospective studies, its use according to the conventional schedule has been associated to high risk of leukemic transformation and second neoplasms, as with other alkylating agents. Consequently, BU is currently used by many hematologists as second-line treatment in Bcr-Abl negative MPN that are resistant to or developed side effects from HU, principally in elderly patients, for whom therapeutic options remain limited 4-6 . The conventional schedule provides a starting dose of 14 mg/week up to obtain a complete haematological response (CHR), defined according to the European LeukemiaNet (ELN) criteria 7 .
In the current retrospective study, we wanted to share our experience with use of long-term and low-dose of BU in elderly patients with ET who are resistant or intolerant to HU.
From 1995 to 2015 we collected 348 patients who received ET diagnosis at our department, according to WHO criteria 1 . The principal clinical and hematological features of entire cohort of ET patients were shown in Table 1. In this population we identified 30 patients who received a second-line treatment with an alternative schedule of BU, defined by low-starting dose (4-6 mg/ week) down to obtain CHR, followed by a dose deescalation of BU over time up to the minimal dose required for the maintenance of CHR. The dose deescalation of BU was performed every 4 weeks and the minimal dose to maintain the CHR was generally between 2 mg/week and 2 mg/month.
We analyzed efficacy, toxicity and risk of leukemic evolution in these settings of patients. Intolerance and resistance to HU were categorized according to ELN criteria for ET 8 . Non-parametric tests, such as Mann-Whitney, Pearson Chi-square and Fischer's exact tests, were used for statistical analysis of continuous and categorical variables. Leukemia-free-survival (LFS) curves were calculated by Kaplan-Meier method and compared with Log-rank (Mantel-Cox) test.
Twenty-six of 30 patients were evaluable for full demographic, diagnostic and therapeutic information and were considered for statistical analysis. As reported in Table 1, we observed some significant differences between BU patients and the entire series of ET patients with respect to age and blood counts at diagnosis. The BU patients were older at diagnosis and presented higher platelet (PLT) count and lower levels of hematocrit (Ht), as compared to global ET population. These demographic At the time of analysis, 16 (61.5%) patients stopped BU: 3 for grade ≥2 hematological toxicity, 4 for disease progression, 3 for drug extra-hematological intolerance or resistance, 2 for grade ≥2 infectious complications; the remaining 4 not for drug-related raisons (principally due to personal patient's decision). Fourteen of 16 patients (87.5%) stopped BU treatment after achievement of CHR.
With a median follow-up of 130 months (range 16-333.3) from diagnosis and 72,7 months (range 5-105.2) from BU start, 6 (23%) patients have died. Causes of death were disease progression in 3 patients (n = 2 leukemic evolution, n = 1 MF transformation), infections (n = 1) and unknown (n = 2). Thrombotic complications after BU start were observed in 5 (19.2%) patients (3 arterial and 2 venous thrombosis). All patients received anti-platelet therapy associated to BU cytoreduction, according to standard guidelines of treatment for ET 2 . No major bleedings or second neoplasms were recorded. Leukemic evolution was observed in 2 (7.7%) patients after a median time of 31.5 months from BU start and 104.8 months from diagnosis. Despite the demographic and biological differences in disease aggressiveness of BU patients, this incidence of transformation into acute leukemia was similar to the entire series of ET patients (2.6%, P = 0.14). The LFS at 15 years was comparable in 2 groups (97 vs. 79.7% at 15 years; Fig. 1, P = 0.12).
In conclusion, to our knowledge, the present study represents the first experience with an alternative longterm and low dose administration of BU in elderly patients with ET who are resistant or intolerant to HU. This different schedule seems to be safe and effective. It provides a high rate of hematological response (92.3% of patients obtained CHR) with acceptable hematological and extrahematological toxicity. The achievement of CHR is rapid (6.5 months) and sustained over time (43.2 months). The risk of leukemic transformation seems to be limited and similar to that is reported in recent literature 4,5 , considering that these patients are very old and previously treated 3,4 . Additional data from larger retrospective multicenter studies or prospective series should be further obtained to confirm the long-term safety and efficacy of this alternative schedule of BU in ET patients.