Mediterranean Journal of Hematology and Infectious Diseases Efficacy and Safety of Cladribine: Subcutaneous versus Intravenous Administration in Hairy Cell Leukemia Patients We Retrospectively Collected Data from Hcl Patients Treated at the National Cancer Institute and Its Affiliated Center, Nasser

Competing interests: The authors have declared that no competing interests exist. Abstract. Cladribine induces durable complete remission (CR) in approximately 85% of hairy cell leukemia (HCL) patients. In Egypt, cladribine is mainly used as IV continuous infusion at a dose of 0.1 mg/kg/day for 7 days and as SC bolus injection at a dose of 0.14 mg/kg/day for 5 days. We aimed to compare the outcome and toxicity between these two regimens. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Introduction. Hairy cell leukemia (HCL) is a rare indolent B-cell leukemia. It accounts for approximately 2% of all leukemias and is characterized by male preponderance (M:F ratio is 4:1). The majority of patients are usually diagnosed over the age of 40 years. Main symptoms are related to cytopenia and/or splenomegaly. 1,2 The diagnosis is confirmed by bone marrow evaluation with flow cytometry. A panel of

Introduction.Hairy cell leukemia (HCL) is a rare indolent B-cell leukemia.It accounts for approximately 2% of all leukemias and is characterized by male preponderance (M:F ratio is 4:1).The majority of patients are usually diagnosed over the age of 40 years.Main symptoms are related to cytopenia and/or splenomegaly. 1,2The diagnosis is confirmed by bone marrow evaluation with flow cytometry.][3][4] Annexin A1, which is not expressed in any other small B-cell lymphoproliferation, is currently the most specific HCL marker. 1,5Tiacci et al have identified BRAF-V600E mutation as the disease-defining genetic event among patients with classical HCL. 6,7ndications for treatment are the presence of constitutional symptoms, symptoms due to splenomegaly, significant cytopenia (absolute neutrophil count (ANC) <1x10 9 /L, platelet count < 100 × 10 9 /L, or hemoglobin level <10g/dL), recurrent infections and autoimmune complications.Main treatment objectives are to control symptoms, normalize blood counts, and achieve lengthy remission.Over the years treatment options have evolved from moderate success with low-dose chlorambucil, splenic irradiation or even anthracyclines, through splenectomy and IFN-α, and eventually to the current treatment of choice: the purine nucleoside analogues pentostatin and Cladribine. 1,2xcellent outcomes are obtained with purine analogues and true primary resistance is seldom observed. 1,2,8,9CR can be achieved in 80-85% of patients and more than 90% of patients are alive at 10 years. 10,111][12][13][14][15] Second CRs can be obtained in more than 75% of patients using the same agents. 12,15ladribine can be administered via intravenous (IV) or subcutaneous (SC) route.7][18][19][20] Alternative treatment schedules such as short daily or weekly infusion. 8The SC injection produces equal (100%) cladribine bioavailability as IV. 21,22Cladribine is the first treatment choice due to its convenient administration schedules that seems to be more advantageous to both patients and physicians. 1,2t the NCI-Cairo, Egypt, the two commonly used regimens of cladrabine are IV continuous infusion at a dose of 0.1 mg/kg/day for 7 days 10 and SC bolus injection at a dose of 0.14 mg/kg/day for 5 days. 23In this study, we compared outcome and toxicity among HCL patients treated with cladribine by SC bolus injection to patients treated with clardribine by continuous IV infusion.

Patients and Methods.
Patients.We retrospectively collected data of HCL patients treated with clardribine during the period 2003 and 2010 at the National Cancer Institute (NCI)-Cairo and Nasser Institute, an NCI-Cairo affiliated center.Eligible patients could have been treated with single or two courses of cladribine.The study was approved by the Ethics Committee of the NCI-Cairo.The diagnosis of HCL was done based on hematopathology review of the peripheral blood, bone marrow, and/or splenic tissue in splenectomized patients as well as the characteristic immunology profile of hairy cells in the bone marrow or the peripheral blood. 3,4The analysis of BRAF mutations was not done to any of the patients as it is not part of our standard of care.Patients should have received clardribine either by IV (IV group) or SC route (SC group) as previously described. 10,23edical records were reviewed for date of diagnosis, gender, age, performance status, initial complete blood count, bone marrow aspirate and biopsy results, organ involvement, immunophenotyping, prior therapies, route of cladribine administration, treatment outcomes and toxicities, last date of follow-up, disease and patient status at last follow-up, date of disease progression, and date and cause of death.
Treatment response and toxicity.Cladribine was given at a dose of 0.1 mg/kg/day for 7-days by continuous IV infusion to 18 patients and at a dose of 0.14 mg/kg/day for 5-days by SC injection to 31 patients.The use of either the IV or the SC route was approved during the period of the study and patients in each group were treated concurrently.The choice of the regimen was according to the discretion of the physician.
Complete and partial remissions are assessed when patients are recovered from cytopenia (usually between month 3-6 post-treatment).Responders (CR or PR) and non-responders are defined according to the Consensus Resolution. 1,24,25In our experience, clearance of hairy cells from the bone marrow might take 4-6 months.The same length of time is sometimes required to document normalization of the size of involved spleen, liver and lymph nodes.So, evaluation of response between the 4th and 6th months of treatment is critical to decide if a second treatment course is needed, in particular if massive organomegaly or severe cytopenia still exists. 25Patients are then followed every three months for the first year and then annually by peripheral blood counts only.
The median and range of blood values at day 10 (nadir) were compared between the two groups.Hematologic and non-hematologic toxicities are graded according to the Common Terminology Criteria for Adverse Events (CTCAE) Version 4.0.
Statistical methods.Overall survival (OS) was defined as the time elapsed between the start of cladribine treatment and death or loss to follow up.Event free survival (EFS) was defined as the time elapsed between the start of cladribine treatment and progression, relapse, second tumor, or death, whichever occurred first.Time to event endpoints was analyzed using Kaplan-Meier method.Comparison between the two groups (subcutaneous versus intravenous route) was performed using log-rank test for survival endpoints, Student t test for continuous variables and Fisher's exact test for categorical variables.
A univariate analysis was conducted in order to assess the prognostic value of the following baseline characteristics on EFS: age (>= 50 versus <50 years), gender, lymphadenopathy and/or hepatomegaly (yes versus no), previous treatment (nil, chemotherapy, Interferon, splenectomy,) and route (SC versus IV).The prognostic value of response to treatment at 4 months on EFS (complete versus partial response) was evaluated using landmark analysis with a landmark time of 4 months.These prognostic factor analyses were performed using a Cox proportional hazards model and Wald test was used to assess the statistical significance of the factors.However, the low number of events for EFS should be considered in the interpretation of the results.
All tests were performed at the 5% two-sided significance level.The statistical analysis was conducted using SAS® version 9.4.

Results.
Clinical characteristics of patients.The relevant characteristics of patients in each treatment group are compared in Table 1.A total of 49 patients were identified: 18 patients in the IV group and 31 in the SC group.Forty-one patients were newly diagnosed and 8 were previously treated.Previous treatment was chemotherapy (n=4), interferon alpha (n=2) and splenectomy (n=2).The median age was 47, range 26-74 years and 78% were males.The clinical characteristics were overall well balanced for patients across the IV and SC groups.Responses to cladribine.After treatment with cladribine, CR was achieved in 94% of patients (17/18)  in IV group and in 97% of patients (30/31) in the SC group (Table 2A).At first evaluation post-treatment (months 3-4), 12 patients were still in PR: two in the IV and 10 in the SC group.At the subsequent evaluation, in the IV group one patient achieved CR and the other died of disease progression and infection neutropenia.Whereas, in the SC group, 9 patients could reach CR, 5 with no further therapy and 4 following a 2nd cycle of clardribine.The tenth patient progressed after the 2nd cycle of clardribine and was lost to follow-up.This patient was previously treated with combination chemotherapy and presented with splenomegaly and lymphadenopathy.
We also did the same analysis restricted to the subset of patients who received cladribine at onset of disease and with single cladribine course in the IV and SC group.CR rates were not different between the two groups, but again time to CR was longer in the SC group.CR rates at week 12 and at subsequent evaluation were 64% and 100% in the SC group, while it was 93% at the two time points in the IV group (Table 2B).
Toxicities.Blood values at diagnosis and at day 10 (nadir) and non-hematologic complications are presented in Table 3A and B. At day 10, there was no difference in the changes in the median blood values between the two groups.Neutropenia G3-4 in the IV group and the SC occurred in 67% (n=12) and 87% (n=27) of patients, respectively.Mucositis was significantly more in the IV group compared to the SC group: 67% (n=12) vs 32% (n=10) P <0.001, respectively.Three patients in the IV group had G3 mucositis and all other patients had GI-II mucositis.
Infections occurred mainly within 10 weeks after treatment.In most patients, the infectious agent couldn't be identified.However, herpes simplex infection was documented in 6 patients (5 in the IV group and 1 and the SC group).In each group, one patient suffered from fungal infection.Fever without any identified source of infection was observed in 4 patients in the IV and in 7 in the SC group.
In the SC group and following the 2nd cycle of treatment one patient had mild bilateral foot drop and one had mild shoulder tendinitis.
Other non-hematologic toxicities were mild (G1-2) and in the form of nausea, vomiting, elevated liver enzymes, cough, headache and bony pains.
Late toxicities included heart failure in one patient in the IV group and second cancer in one patient in the SC group.Causes of death in the IV and SC groups are summarized in Table 3C and are detailed below.
Event Free Survival.The median time to follow up was 33.5 months (range, 10-28 months).Median EFS were 53 and 63 months in the IV and the SC group, respectively.The 3-year EFS was 60% and 96% in the IV and the SC group, respectively (p=0.104) (Figure1).
In addition to the patient died shortly after     heart disease.
In the SC group, one patient died of disease progression and two experienced relapsed at 44, 13 and 37 months, respectively.Both relapsed patients were re-treated with SC clardribine, one achieved PR and was lost to follow -up, and the second obtained CR.Additional patient was diagnosed with diffuse large B cell lymphoma 70 months from the initiation of therapy.This patient died during treatment due to tumor lysis syndrome.
Clinically relevant prognostic factors that would influence the EFS were analyzed and the results are shown in table 4. Results suggests that having lymphadenopathy and/or hepatomegaly is of poor prognosis (HR=6.7,95% CI= 1.3-33.9,p=0.023).No other factors, including route of administration, showed a significant prognostic effect in this analysis.The analysis of EFS by response to treatment did not show any difference between patients in CR and patients in PR at months 4 (p=0.925)(Figure 2).
Overall survival.Median OS was 74 months in the IV group and was not reached in the SC group.The 3-year OS was 81% and 100% for patients in the IV and SC group, respectively (p=0.277) (Figure 3).Discussion.To our knowledge, this study that compared HCL patients treated with SC cladribine vs IV cladribine is the first from a developing country.We reported the clinical features and the outcome of 49 Egyptian HCL patients treated with cladribine in the period between 2004 and 2010.The clinical presentation of patients in this series is similar to the literature. 1 The results show that SC cladribine is safe and produces a remission rate similar to the IV route.Importantly, the two regimens resulted in high EFS.The SC route was associated with non-significant but important longer DFS (median 63 vs 53 months and 3year EFS 96% vs 60%).EFS than the IV regimen.
The overall CR rate of this series was 96% which matches well to CR rates reported in previous studies which ranged between 87%-100%. 1,2Interestingly, the time to achieve CR was longer in the SC group (table 2 A and B), and 5 patients required a second course of cladribine to reach CR.While 89% of patients were in CR at first evaluation in the IV group, only 68% of the SC group reached CR.At the subsequent assessment the rates improved to 94% and 97%, respectively.Studies investigating the efficacy of the 7 day and 5 day regimens of SC cladribine reported CR rates of 81% and 76%, respectively. 23,26Therefore, in this study, the estimated CR rate is very encouraging for SC cladribine.
Cladribine is a generally well-tolerated 1,2,23,26-29 drug, and cytopenia and culture negative fever are the most commonly expected side effects. 1,2,27Infections are also a common toxicity and occur mainly during the first weeks of treatment along with cytopenia. 1,2,27n a review by Maevis et al, the frequency of fever among HCL patients treated with cladribine ranged between 40-69%, and not all fevers were associated with signs of infection. 2The authors suggested that fever may be a sign of cytokine release in HC rather than infection in some patients. 2In accordance, the most frequent toxicity in this study was a myelosuppression (approximately 80% of patients).Fever (with or without signs of infection) was seen in 35% of all patients, and infections were diagnosed mainly in the first 10 weeks post-therapy.Nevertheless, infectious episodes (mainly herpes simplex) and mucositis were more frequent in the IV group.
A number of investigators have indicated that the risk of developing second malignancy is higher in HCL patients. 18,30Whether this is related to common genetic susceptibility or due to treatment exposure is not known.In a study of 358 patients with HCL treated with cladribine, 8% (27/358) developed second tumors. 18According to the Surveillance, Epidemiology and End Results (SEER) data, of 3104 patients with HCL, the incidence of second cancer was 32% compared to 23% in the general population (SIR=1.2;95% CI 1.1-1.4 for all, 6.6 for Hodgkin lymphoma, 5.0 for non-Hodgkin lymphoma and 3.6 for thyroid cancer. 30In our study, one patient (2%) died of diffuse large B cell lymphoma 70 months after CDA therapy.Without longer follow up, analysis of late effects such as second cancers will not be valuable.
In a recent literature review, the 5-year PFS of patients treated with cladribine was found to be in the range of 72-84%. 2 In correspondence to this range, the overall 3-year EFS of patients in this report was 77%.However, patients in the IV group experienced more failures than patients in the SC group (3-year EFS was 60% vs 96%) and median EFS was shorter in patients in the IV than in the SC group (53 vs 63 months, respectively).This trend did not reach statistical significance in the univariate analysis.A number of studied evaluated SC cladribine as alternative to the IV route.In one study no relapse was seen after 20 months follow up of 73 HCL patients treated with the 7 day SC schedule. 26In the second, 62 patients received the 5 day SC regimen and the 18 months-EFS was 68%. 23A comparison between the different results is not possible due to variability of end point definitions and methods on analysis.
Different prognostic factors have been previously described, such as cytopenia 1,12,13 and lymphadenopathy at presentation 14,31 and response to purine analogues. 12,14,25,29Two main studies indicated response to purine analogs as the most significant factor for DFS. 12,25Therefore we compared EFS between patients in CR and patients in PR and no statistically significant difference between the two subgroups was found.
Albeit non-significant, the 3-year OS for patients treated with the SC regimen was 100% compared to 81% in patients treated with the IV regimen.This is very supportive to the effectiveness of the SC regimen.Main causes of death in the IV group were complications related to either disease progression or treatment re-challenge.
Being a rare condition, conducting prospective studies in HCL is a real challenge.Limitations to retrospective studies include heterogeneity in data recording and data collection as well as variability in treatment decisions among centers.However, data collected from single center are usually more homogenous as is the case of the current series.Therefore, this study would provide valuable information on such rare disease.
In conclusion, this is the first report of a large series of Egyptian HCL patients comparing the SC to the IV cladribine regimen.Our data confirms that SC cladribine is an excellent alternative to the IV regimen.SC cladribine is well tolerated and is associated with favorable EFS.Together with the convenience of the SC route, this regimen should be recommended as of choice to treat HCL patients in institutes with limited resources such as those in developing countries.

Figure 1 .
Figure 1.EFS of the IV and SC cladribine group.

Figure 2 .
Figure 2. Event Free Survival, by response at month 4 (CR versus PR) -Landmark analysis.Months counted from the landmark time of 4 months.

Figure 3 .
Figure 3. OS of the IV and the SC group.

Table 1 .
Baseline characteristics: overall and by treatment group (IV versus SC).
(*) For the comparison between the IV and SC groups (Fisher's exact test or t-test).

Table 2A .
Response: overall and by treatment group (IV versus SC).
(*) For the comparison between the IV and SC groups (Fisher's exact test).

Table 2B .
Response in subset of patients treated at onset of disease and with a single cladribine course: overall subset and by treatment group (IV versus SC).

Table 3 .
Cladribine toxicity in the IV and the SC.

Table 3A .
Hematologic complications the IV and the SC group.

Table 3B .
Non-hematologic complications the IV and the SC group.

Table 3C .
Causes of death in the IV and the SC group.

Table 4 .
Prognostic factor analysis of Event Free Survival (EFS).