• hairy cell leukaemia;
  • vemurafenib;
  • BRAF V600E

Following the report that 100% of classical Hairy Cell Leukaemia (HCL) patients have the BRAF V600E mutation (Tiacci et al, 2011), we provided early independent confirmation of this result on a cohort of Cambridge patients (Boyd et al, 2011). The mutation is readily targeted by the small molecule inhibitor vemurafenib, which has now been used in a dose escalation protocol to treat a single patient with relapsed HCL (Dietrich et al, 2012). In this report, we document our successful experience treating a patient with relapsed biallelic BRAF V600E mutated HCL. We believe this is the first report of non-escalated low dose vemurafenib therapy in HCL.

From our published cohort of 48 HCL patients (Boyd et al, 2011), we identified a single 68-year-old male patient with a biallelic mutation. The patient initially presented with HCL in the mid-1990s when he was managed with splenectomy followed by cladribine therapy in 1997. He remained in remission for over 10 years, relapsing in 2008, when he was retreated with cladribine. Retrospective BRAF sequencing analysis from archived material from 1996 to 2008 confirmed he had a monoallelic mutation in 1996, but at relapse in 2008, he had acquired a biallelic BRAF V600E mutation (data not shown). He relapsed again, requiring 3rd line chemotherapy (six cycles of pentostatin / rituximab) in 2011, achieving a partial bone marrow remission with normalization of peripheral blood counts. Unfortunately his disease relapsed within 6 months, with worsening cytopenias and increasing numbers of circulating hairy cells. The patient became dependent on red cell/platelet transfusions along with intermittent granulocyte colony-stimulating factor (G-CSF) therapy. He was intolerant of interferon-alpha.

The patient became increasingly unwell with lethargy, flu-like symptoms, sweats, general debility and pelvic/lumbar pain. As there were no further therapeutic options available we applied for dispensation to treat the patient with vemurafenib. In the 3 months prior to treatment he had received 14 units of red cells and 9 units of platelets and quality of life was poor. A pre-treatment bone marrow trephine biopsy revealed complete replacement of normal architecture with HCL (Fig 1A) and the baseline magnetic resonance imaging (MRI) scan of pelvis and femora revealed diffusely abnormal homogenous bone marrow signal with increased T2 fat suppressed (T2 fs) signal (Fig 1D) and decreased T1 signal. His pre-treatment peripheral blood showed a prominent hairy cell leucocytosis, which could be quantified by flow cytometry (Fig 1). Treatment was commenced with vemurafenib 240 mg BD and the dose was continued with no breaks or escalation for 58 d. Treatment was well tolerated. The only patient-reported side effect was the growth of three wart-like skin lesions, which appeared within 3 weeks of starting therapy. Surgical removal confirmed benign seborrhoeic keratosis. Early growth of squamous skin lesions has been observed as a common side effect in melanoma patients on vemurafenib therapy, and surgical removal is standard practice (Mattei et al, 2012).


Figure 1. Panels A to C represent paired peripheral blood flow cytometry plots with Haematoxylin & Eosin-stained bone marrow trephine samples taken on the same day: (A) pre-treatment, (B) day 37 on vemurafenib, (C) 60 d post-completion of vemurafenib. The flow plots depict all peripheral blood leucocytes with clear separation of the normal CD20+ B cells (blue) and CD20+/CD103+ hairy cells (orange). On therapy there is a clear reduction in bone marrow infiltration of Hairy Cell Leukaemia (HCL) with a restoration of normal haematopoiesis. However, following 60 d without therapy, there is an increase in the bone marrow infiltration with HCL, which is mirrored by the increase in peripheral blood HCL demonstrated by flow cytometry. (D). Pre-treatment coronal T2 fat-supressed magnetic resonance imaging (MRI) of pelvis and femora revealed diffusely abnormal homogenous bone marrow signal with increased T2 fat suppressed signal compared to normal signal within the greater trochanteric apophyses. (E). Repeat MRI 9 d after completing therapy showed improvement with recovery of normal bone marrow signal.

Download figure to PowerPoint

The peripheral blood HCL count (as quantified by flow cytometry), haemoglobin, platelet count, neutrophil count and serum bilirubin, over a 165 d period from 47 d pre-treatment to 60 d post treatment are presented in Fig 2. On day 1 of treatment, the full blood count: haemoglobin 109 g/l(supported), total white cells 23·8 × 109/l, hairy cells 19·1 × 109/l, neutrophils 0·47 × 109/l, platelets 20 × 109/l (supported). Within 8 h of the first dose of vemurafenib, the peripheral blood hairy cell count increased to 30 × 109/l, but then decreased from day 5 of therapy. The hairy cell count continued to decrease on therapy to 0·002 × 109/l by the final day of treatment, i.e. a 10 000-fold reduction in peripheral blood hairy cell numbers. The platelet count remained supported for the first 2 weeks of treatment (transfusions on days 3, 9, 14). Thereafter the platelet count rose steadily to a value of 193 × 109/l by day 58 of treatment. The neutrophil count initially increased in the first 4 d of therapy but then fell to between 0·2–0·35 × 109/l despite daily G-CSF injections. Without interrupting treatment, the neutrophil count started to rise by day 19 and the final dose of G-CSF was given on day 28. By the final day of treatment, the neutrophil count was 3·95 × 109/l. The patient received 3 further 2 unit red cell transfusions while on therapy, on days 7, 14 and 22. From day 28 of treatment the haemoglobin started to increase without transfusion support and was 115 g/l by the final day of treatment. We could not convincingly demonstrate an increase in red cell transfusion requirements during the first 3 weeks of therapy. The only biochemical abnormality of note while on treatment was a rise in unconjugated bilirubin, with no significant rise in other liver enzymes, electrolytes or creatinine. The pre-treatment bilirubin was 10 μmol/l (0–17 μmol/l), which increased to 24 μmol/l within 24 h of starting therapy. The bilirubin peaked at 41 μmol/l (85% unconjugated) on day 15 before progressively falling while therapy continued. We could not demonstrate evidence of haemolysis as judged by blood film appearances, normal serum lactate dehydrogenase and a mildly elevated serum haptoglobin. Direct anti-globulin test has remained negative. The bilirubin rose again to 30 μmol/l on the final day of therapy, before returning to normal when treatment stopped.


Figure 2. Plots of serial peripheral blood values over a 165-day period, from 47 d pre-treatment to 60 d post-treatment completion. The broad grey highlight represents the 58 d period of continuous therapy with vemurafenib 240 mg BD. The panels represent the HCL count × 109/l as quantified by flow cytometry; the haemoglobin in g/l with arrows indicating the seven transfusions of 2 units of red cells; the platelet count × 109/l with arrows indicating the 10 transfusions of single adult platelet doses; the neutrophil count × 109/l with the black bar representing ongoing granulocyte colony-stimulating factor (G-CSF) therapy; serum bilirubin in μmol/l.

Download figure to PowerPoint

A bone marrow biopsy taken on day 16 of therapy showed some early recovery of normal haemopoiesis, while a biopsy taken on day 37 of treatment showed clear reduction in leukaemia burden with marked improvement in normal haematopoiesis (Fig 1B), constituting a partial response to therapy. Repeat MRI scan of the pelvis post-therapy confirmed a significant improvement in bone marrow signal (Fig 1E).

Treatment was stopped after 58 d and the patient has now been observed off treatment for 2 months. He remains very well, enjoying an excellent quality of life with a normal performance status and he remains transfusion independent. He has developed no further skin lesions. While he has maintained his haemoglobin and neutrophil count off treatment, his platelet count has fallen to 116 × 109/l at 60 d post therapy. This has coincided with a steady rise in his peripheral blood hairy cell count over the 60 d without treatment to 0·19 × 109/l. A bone marrow biopsy taken 60 d post-completion of treatment also showed an early increase in HCL infiltration compared with a biopsy on day 37 of therapy (Fig 1C). Overall, the features are compatible with early relapse of HCL following treatment cessation, although, as the patient remains well, retreatment with vemurafenib has not yet commenced.

The HCL story has been a remarkable example of ‘bench-to-bedside’ medicine. Within 12 months of the first report of the discovery of the BRAF V600E mutation, a patient had been successfully treated with a targeted inhibitor (Dietrich et al, 2012). In this report we document the first use of 58 d of continuous low dose vemurafenib to successfully induce a partial remission with striking clinical benefit for our patient who had a biallelic V600E mutation. He acquired the biallelic mutation at some stage during an 11-year remission following cladribine monotherapy, but it is not clear at this stage whether the acquisition of a biallelic mutation associates with different disease biology, inferior response to standard treatments or shorter remission duration. In contrast with the report from Dietrich et al (2012), we chose to avoid dose escalation, primarily due to the very limited experience of using this drug in HCL. As our patient was cytopenic pre-treatment, it is not clear whether there was an initial fall in his blood counts while on therapy. Neutropenia is uncommon, but has been reported in melanoma patients receiving vemurafenib (Chapman et al, 2012). Our patient also developed a rapid unexplained, isolated increase in unconjugated bilirubin that initially normalized with ongoing therapy, but had increased again by the final day of treatment. We could not link this finding to haemolysis and the cause of this adverse event remains unexplained. As there are no data regarding ongoing therapy with vemurafenib in HCL, we chose to stop treatment after 58 d in line with the report by Dietrich et al (2012). In the first 60 d off treatment he has continued to enjoy an excellent quality of life. However, we accept that data from his peripheral blood and bone marrow biopsy suggest an early relapse of his leukaemia off therapy and we plan to restart therapy at 240 mg BD when his cytopenias return.

There is much to learn with regards to the adverse event profile of vemurafenib in HCL and the optimal drug dose/duration of treatment remains unknown. We therefore strongly support efforts to establish a clinical trial programme to investigate the role for vemurafenib in HCL. In the interim, we also support plans to document carefully the adverse event profile and response rates of HCL patients treated off trial with BRAF inhibitors, coordinated by Thorsten Zenz and others (


  1. Top of page
  2. References
  • Boyd, E.M., Bench, A.J., van ‘t Veer, M.B., Wright, P., Bloxham, D.M., Follows, G.A. & Scott, M.A. (2011) High resolution melting analysis for detection of BRAF exon 15 mutations in hairy cell leukaemia and other lymphoid malignancies. British Journal of Haematology, 155, 609612.
  • Chapman, P.B., Hauschild, A., Robert, C., Haanen, J.B., Ascierto, P., Larkin, J., Dummer, R., Garbe, C., Testori, A., Maio, M., Hogg, D., Lorigan, P., Lebbe, C., Jouary, T., Schadendorf, D., Ribas, A., O' Day, S.J., Sosman, J.A., Kirkwood, J.M., Eggermont, A.M., Dreno, B., Nolop, K., Li, J., Nelson, B., Hou, J., Lee, R.J., Flaherty, K.T., McArthur, G.A. & BRIM-3 Study Group. (2012) Improved survival with vemurafenib in melanoma with BRAF V600E mutation. New England Journal of Medicine, 364, 25072516.
  • Dietrich, S., Glimm, H., Andrulis, M., von Kalle, C., Ho, A.D. & Zenz, T. (2012) BRAF inhibition in refractory hairy-cell leukemia. New England Journal of Medicine, 366, 20382040.
  • Mattei, P.L., Alora-Palli, M.B., Kraft, S., Lawrence, D.P., Flaherty, K.T. & Kimball, A.B. (2012) Cutaneous effects of BRAF inhibitor therapy: a case series. Annals of Oncology, doi: 10.1093/annonc/mds292.
  • Tiacci, E., Trifonov, V., Schiavoni, G., Holmes, A., Kern, W., Martelli, M.P., Pucciarini, A., Bigerna, B., Pacini, R., Wells, V.A., Sportoletti, P., Pettirossi, V., Mannucci, R., Elliott, O., Liso, A., Ambrosetti, A., Pulsoni, A., Forconi, F., Trentin, L., Semenzato, G., Inghirami, G., Capponi, M., Di Raimondo, F., Patti, C., Arcaini, L., Musto, P., Pileri, S., Haferlach, C., Schnittger, S., Pizzolo, G., Foà, R., Farinelli, L., Haferlach, T., Pasqualucci, L., Rabadan, R. & Falini, B. (2011) BRAF mutations in hairy-cell leukemia. New England Journal of Medicine, 64, 23052315.