“Hairy Cell Leukemia (HCL): ‘Real World’ Outcome”

Abstract

HCL is an uncommon B cell lympho-proliferative disorder with high remission rates. There is paucity of data on the long-term outcome of HCL from India. We retrospectively collected data from individual case records of patients with HCL who were treated in Cancer Institute, Chennai from January 2001 until January 2018. Sixteen patients were diagnosed with HCL and were treated with cladribine (81%), interferon (13%) and one patient received only best supportive care (6%). All the treated patients achieved complete response. More than half of the patients developed febrile neutropenia but there were no treatment related mortality. The 5-year DFS was 77% and 5-year OS was 80%. Relapse of disease was seen in 27%. HCL is a curable malignancy with high remission rates and survival comparable to patient treated in west.

Introduction

HCL is a rare chronic B-cell lymphoid malignancy with an indolent course accounting for 2% of leukemia [1]. It is characterized by progressive accumulation of clonal mature B-lymphocytes with “hairy” projections on the cell surface in the blood, bone marrow, and spleen [1]. The median age at presentation is 55 years with a male preponderance [2]. The common presenting complaints are splenomegaly and infection [3]. Treatment with nucleoside purine analogs like cladribine and pentostatin has shown excellent remission rates and long-term survival [4]. There are limited studies on the long-term outcome of HCL form India. We looked at the outcomes of HCL treated at our center over a 17-year period.

Methods

Data of all consecutive patients of HCL treated in Cancer Institute, Chennai between January 2001 and January 2018 were retrospectively obtained from the individual case records. Clinical symptoms at diagnosis (fever, weakness, abdominal lump, loss of weight, bleeding, presence of infection), ECOG performance status, clinical signs (splenomegaly, hepatomegaly, lymphadenopathy), complete hemogram, peripheral blood smear examination findings and bone marrow infiltration with atypical lymphoid cells with cytoplasmic fine prolongations (hairy cells), flow cytometry of peripheral blood or bone marrow aspirate, types of therapy, complications and relapse rate were monitored. Responses to therapy and survival outcomes were analyzed.

Indications for treatment were symptomatic disease with fatigue, symptomatic splenomegaly or hepatomegaly, weight loss of more than 10% in last 6 months, cytopenias (hemoglobin < 11 g/dL, platelets < 100 × 10⁹/L and/or absolute neutrophil count < 1 × 10⁹/L), progressive lymphocytosis or lymphadenopathy. Cladribine was given at the dose of 0.1 mg/kg/day for 7 days as a continuous infusion through a central venous access or at 0.14 mg/kg/day for 5 days as a 2-h infusion. Antibacterial, antiviral and antifungal and anti-pneumocystis prophylaxes were levofloxacin 500 mg once a day, acyclovir 400 mg twice a day, fluconazole 200 mg twice a day and co-trimoxazole 960 mg three times a week respectively. Prophylaxis was continued for at least 2 months after recovery of blood counts. All blood components except fresh frozen plasma/cryoprecipitate were irradiated before transfusion. All patients remained hospitalized till recovery of neutrophil count. Growth factors and intravenous antibiotics were used in patients who had febrile neutropenia.

Response was assessed according to consensus guidelines for the diagnosis and management of patients with classical hairy cell leukemia [5]. CR (complete response) is near normalization of peripheral blood counts: hemoglobin > 11 g/dL (without transfusion), platelets > 100 × 10⁹/L, absolute neutrophil count > 1.5 × 10⁹/L; regression of splenomegaly on physical examination and absence of morphologic evidence of HCL on both the peripheral blood smear and the bone marrow examination. DFS (disease free survival) was calculated from date of diagnosis to date of relapse, death or date of last follow up. OS (overall survival) was calculated from date of diagnosis to date of death due to any cause or date of last follow up. Survival outcomes were analyzed by Kaplan Meir’s analysis by SPSS version 23.

Results

From January 2001 until January 2018, 16 patients were diagnosed with HCL at our Institute. Baseline characteristics are mentioned in Table 1.

Table 1.

Baseline characteristics

Baseline Characteristics (n = 16)
Median age at presentation	53.5 years (range 38–65)
Male to female ratio	4.3:1
Fever	56%
Splenomegaly	100%
Hepatomegaly	13%
Baseline infections	19%
Bleeding	6%
Median hemoglobin	8.9 g/dl (range 2.5–13)
Median leucocyte count	4.35 × 109/L (range 0.3–28)
Median platelet count	95 × 109/L (range 35–183)
Hairy cells in bone marrow	100%

Clinical features at presentation: The median duration of symptoms was 2.5 months (range 1–36). About 19% has documented baseline infection. Clinically documented infections were pneumonia (13%), urinary tract infection (6%) while 38% had febrile neutropenia with no clinical or microbiological source of infection. Massive splenomegaly (crossing the umbilicus) was seen in 38% patients. The ECOG (eastern cooperative oncology group) performance status was 1, 2, and 3 in 69%, 19% and 12% respectively.

Laboratory parameters at presentation: Leukocytosis was present in 31% of the patients. Pancytopenia was present in 38% of the patients. TRAP positivity was seen in 50%, and 56% cases showed classical hairy cell in peripheral smear. By flow cytometry (FCM) analysis, CD19, CD20 and CD22 were positive in 100%, CD103 in 81%, and CD11c, CD25 and FMC7 in 72% of patients.

The median duration of follow-up was 60 months. Patients were treated with cladribine 81% (n = 13), or interferon 13% (n = 2), while 1 patient received only supportive care. Nine patients received cladribine as conventional 24-h infusion for 7 days, while 4 patients received cladribine as short daily infusion for 5 days. Among patients who were treated (n = 15), CR was seen in all patients (100%). Seven patients (53%) who received cladribine developed grade 3/4 febrile neutropenia, requiring growth factors and broad spectrum intravenous antibiotics. Of seven patients, one patient had MDR Klebsiella sepsis, one patient had pneumonia and one patient had urinary tract infection. Four patients had no clinical or microbiological source of infection. There were no treatment related mortalities. None of the patients developed nephrotoxicity with cladribine. Median time to clinical and hematological response was 2.9 months (range 1–4.8). Median DFS (disease free survival) and median OS (overall survival) were 51.6 and 56.7 months respectively. The 5-year DFS was 77% and 5-year OS was 80%.

Relapse rate was 27% (cladribine treated = 3: interferon treated = 1). Median time to relapse in relapsed patients was 60 months (range − 49 to 131 months). One patient who relapsed after cladribine, was treated with rituximab and the other was treated with rituximab followed by cyclosporine and prednisolone and both of them achieved remission. The patient who relapsed after interferon, did not receive further treatment due to poor performance status. One patient didn’t receive full dose of cladribine as he developed febrile neutropenia and urinary tract infection on day 4 of cladribine; he had residual disease but achieved CR after 8 cycles of Rituximab. There was no second malignancy in any of the treated patients.

Discussion

Hairy cell leukemia is an indolent disorder, which has the potential of attaining long-term cure. The other studies of HCL from India are listed in Table 2. The median age at presentation was 53.5 years, which is similar to a recent Indian study of 50 years by Somasundaram et al. [6] but significantly younger as compared to the western literature. In this series, 19% of patients had baseline infection (e.g., pneumonia, urinary tract infection) similar to 23% by Somasundaram et al. [6], but higher compared to western literature. Active infection should be treated prior to initiating treatment with standard dose purine analogs. About 38% presented with pancytopenia compared to 30% in Abraham et al. [7] and 28% in Somasundaram et al. [6]. In 11 out of 16 patients, diagnosis was based on immunophenotyping. Majority of the cases showed expression of the characteristic markers of HCL e.g., CD11c, CD25, CD103, CD123, FMC-7, along with other B cell markers. Around 30% of the patients had leukocytosis and CD 25 negativity, possibly HCL variants. The variants have immunophenotype differences, lack of BRAF mutation and poor response to therapy [8]. CD200 can be used to differentiate HCL from splenic diffuse red pulp lymphoma and splenic marginal zone lymphoma [9].

Table 2.

Studies of HCL from India

Study	Center	N	Complete remission %	Relapse %	Median DFS (months)	Median OS (months)
Current study	Cancer Institute, Chennai	16	100	27	51.6	56.7
Abraham et al. [7]	KIdwai Institute, Bengaluru	10	88	20	54	61
Somasundaram et al. [6]	AIIMS, New Delhi	35	93	18	23	26
Lad et al. [12]	PGI, Chandigadh	24	87.5	16.6	NR	NR
Galani et al. [21]	TMH, Mumbai	28	90	5	NR	NR
Medhi et al. [22]	AIIMS, New Delhi	23	84	11	NR	NR
Gupta et al. [23]	PGI, Chandigadh	21	75	NR	NR	NR
Chatterjee et al. [24]	AIIMS, New Delhi	15	80	NR	NR	NR

NR not reported

CR (Complete Response) rate was 100% in our study, slightly higher compared to other Indian studies (80–90%). The role of minimal residual disease (MRD) is controversial in HCL and many patient with MRD (minimal residual disease) positivity do not relapse even many years after therapy [10]. Moxetumomab pasudotox is a non-chemotherapy regimen that can eliminate MRD in a significant percentage and enhance the CR duration [11]. Relapse rate in our study was 27%, slightly higher compared to other studies (20% in Abraham et al. [7] and 18% in Somasundaram et al. [6]).

This is the fourth study (after Somasundaram et al., Abraham et al. and Lad et al.) detailing the outcome of hairy cell leukemia reported from India [6, 7, 12]. Median time to response was 2.9 months in our study, which is similar as other studies. Our study confirms previous observations that cladribine administered to patients with HCL induces high CR rates. About 53% developed grade 3/4 neutropenia, which was managed with growth factors and antibiotics. Recently it has been found that cladribine can be delivered as short 2-h infusion with similar efficacy and safety [13]. 14 patients in our study received cladribine as short infusion. We found that it was equally safe and as effective as 24-h infusion. Median DFS (51.6 months) & OS (56.7 months) were comparable to study by Abraham et al. [7] (54 months and 61 months respectively); higher than Somasundaram et al. [6] (23 and 26 months respectively). A population based analysis from Netherlands has shown that HCL patients can have a normal life expectancy [14].

A meta-analysis by Andrasiak et al. [15] showed that the most effective treatment for HCL was cladribine followed by rituximab maintenance. Investigational non chemotherapy options include moxetumomab pasudotox [16], which targets CD22; vemurafenib or dabrafenib, which target BRAF V600E protein: trametinib [17], which targets mitogen-activated protein kinase enzyme (MEK); and ibrutinib, which targets Bruton tyrosine kinase (BTK) [18]. BRAF mutations are present in more than 80% [19] of patients with HCL and these patients present at a younger age and are less response to cladrabine [20]. HCL variant type can have mutations targeting KDM6A that target encode a lysine demethylation protein which can permit development of personalized therapy in the future [19]. Though limited by its retrospective nature and small number of patients, our study confirms that cladribine is effective and can be safely delivered with appropriate precautions. Majority of our patients were treated in “general wards” along with 10–15 other patients with no access to isolation or focused nursing and hence this data reflects a “real-world” situation existing in most of the major centers in the country.

Conclusion

HCL is a curable malignancy and high remission rates and survival comparable to patients treated in the west. Cladribine produced remarkably high remission rates with single course and can be safely delivered with appropriate precautions.

Funding

The study was supported by Cancer Institute (WIA) funds.

Compliance with Ethical Standards

Conflict of interest

The authors declare that they have no conflict of interest.