Introduction

Hairy cell leukemia (HCL) is a chronic B-cell lymphoproliferative disorder characterized by circulating mononuclear cells and clinically manifested by pancytopenia and splenomegaly or other extramedullary organ involvement.^{1, 2} Patients (pts) with this relatively indolent disease eventually require treatment for life-threatening pancytopenia or symtomatic splenomegaly. The first introduction of 2-chlorodeoxyadenosine (2-CdA), an adenosine deaminase (ADA)-resistant purine analog,³ in the treatment of HCL by Piro⁴ represents a milestone in modern hematology comparable to ATRA in acute promyelocytic leukemia because 2-CdA is highly effective in B-cell lymphocytes, particularly hairy cells. Following these encouraging results, several studies have reported a long-term follow-up of 3–5 years after treatment with 2-CdA.^{5, 6, 7, 8, 9}

In this paper, we analyze the initial and long-term results at 10 years, such as clinical outcome, duration of response, toxicity and survival of 44 HCL pts uniformly treated with a single 7-day course of 2-CdA by continuous infusion. The experience of both pretreated pts, including splenectomy, interferon (IFN) and deoxycoformycin (dCF), as well as previously untreated pts will be described. Also, the possibility of obtaining another response using this purine analog in pts who relapsed after 2-CdA treatment will be addressed. Owing to frequent bone marrow evaluations by various morphologic techniques following treatment, new information as to the dynamics of hairy cell clearing from the bone marrow could be accumulated.

Patients and methods

Patient population

Between August 1990 and July 1997, 44 consecutive pts with HCL were treated at the Department of Hematology/Oncology at the University of Munich. In all, 34 pts were males and 10 were females, with a median age of 57 years (33–77) at the time of initiation of 2-CdA treatment. Altogether, 11 pts were previously treated with either splenectomy (n=6), IFN (n=9), or dCF (n=3), or all procedures in sequence (n=5). The median time from first diagnosis until the start of 2-CdA treatment in these pretreated pts was 47 months (13–160). The median time from first diagnosis of HCL in all 44 pts until the start of treatment with 2-CdA was 4 months (0–160). At the start of 2-CdA treatment, 34 pts had a variable degree of splenomegaly, two of them in addition lymphadenopathy or hepatomegaly, three had no palpable spleen. From the six pts previously splenectomized, one presented with osseous and one with lymph node involvement. Five pts displayed a leukemic blood picture with high white blood cell counts. In these pts, splenic lymphoma with villous lymphocytes could be excluded. For diagnosis of the various morphologic subtypes, the histopathologic criteria for classification were employed, published elsewhere.¹⁰ At diagnosis, 24 pts showed histomorphologically an ovoid subtype of HCL, 12 indented, two mixed ovoid-indented, three convoluted and three remained unclassified.

All pts had a physical examination at entry into the study as well as at each follow-up. They also received routinely a two-dimensional ultrasound of the abdomen to exclude or establish hepato- or splenomegaly or other extramedullary disease at all times of presentation. In case of lymph node enlargement, a total body CT scan was performed. Detailed clinical characteristics of the pts are summarized in Table 1.

Table 1 - Clinical characteristics.

Full table

Eligibility

Eight of the 44 pts presented with a second malignancy before start of treatment with 2-CdA (Ca of the bladder, cervix, prostate gland, breast, colon and tongue). Eligibility criteria included the following: (1) morphologically confirmed diagnosis of HCL based on the bone marrow aspirate, core biopsy and peripheral blood smear obtained within 4–6 weeks of study entry; (2) evidence of active disease including one or all of the following: neutropenia (neutrophil count <1500/l), anemia (hemoglobin level <12 g/dl), thrombocytopenia (platelet count <100 000/l), symptomatic splenomegaly or adenopathy; (3) no active infection; (4) no prior treatment for HCL within 4 weeks of receiving 2-CdA; (5) pts should not have received prior treatment with 2-CdA; (6) normal renal and hepatic function as evidenced by a serum creatinine of <1.5 mg/dl and a serum bilirubin of <2 mg/dl. In all 44 pts, peripheral blood flow cytometry was performed to confirm the diagnosis before treatment with 2-CdA. Immunofluorescence studies included the use of monoclonal antibodies against CD 5, CD 11, CD 19, kappa and lambda chains, CD 20, CD 22, CD 25, HLA determinants and CD 103. Tartrate-resistant acid phosphatase (TRAP) was used as a cytological marker.

Treatment

All pts were treated according to a Group C protocol of the National Cancer Institute (NCI), protocol number 218321. They received a single cycle of 2-CdA (Leustatin; Cladribine; Ortho-Biotech-Cilag, Germany) at a dose of 0.1 mg/kg/day by continuous intravenous infusion for 7 days. The vials were stored at -20°C, and after thawing, diluted in 500 ml of 0.9% sodium chloride for infusion.

Supportive care

All pts were hospitalized and treated in a single bedroom during hypoplasia until hematologic recovery. No antibiotic prophylaxis was given. Broad-spectrum antibiotics were administered when fever exceeded 101°F. When blood cultures remained sterile, and no clinical evidence of infection could be substantiated, antibiotics were stopped. Complete blood counts with differentials and chemistry panels were performed three times weekly. Four pts received G-CSF because of prolonged neutropenia (37, 40, 55, 74 days). One of these pts had suffered from Ca of the cervix prior to the diagnosis of HCL, and was treated by pelvic irradiation (74 days); one pt experienced FUO WHO IV following 2-CdA (55 days); one pt had received interferon for 3 years prior to treatment with 2-CdA (40 days); one pt had several episodes of FUO following treatment with 2-CdA.

Response criteria

Pts were evaluated for response when hematological recovery occurred, as a rule, 2–4 weeks after the end of treatment. This included bone marrow aspirate and core biopsy, and complete hemogram including differential blood count. On both, bone marrow and blood smear, the cytochemical staining for tartrate-resistant acid phosphatase was mandatory in addition to the regular Giemsa staining, also immunofluorescence staining for CD 103. These procedures were repeated every 3 months thereafter during the first year, every 6 months during the second year, and annually thereafter. In addition, the spleen size was monitored by ultrasound in two diameters at the same intervals.

According to the NCI guidelines based on a consensus resolution,¹¹ all of the following criteria had to be met to be considered a complete response (CR): normalization of hemoglobin (12 g/dl), absolute neutrophil count (1500/l), platelet count (100 000/mm³), no hairy cells in the peripheral blood, no morphological evidence of hairy cell infiltration on bone marrow biopsy or aspirate, absence of disease-related symptoms and disappearance of all measurable disease. Immunophenotypic analysis of peripheral blood and bone marrow aspirate and marrow immunohistochemical stainings were not included in the definition of CR. However, both techniques were routinely employed during follow-up for detection of minimal residual disease and relapse. CR with residual disease (CR-RD) was defined as for CR, but with persistence of 1–5% hairy cells in the marrow detected by immunocytology and/or immunohistology, but no circulating hairy cells. Partial response (PR) was defined as meeting the peripheral blood count criteria for CR/CR-RD with 5–10% residual hairy cells in the marrow and correction of at least one cytopenia without decrease in any cell count; reduction in palpable abnormalities on physical examination by more than 50% and no hairy cell in the peripheral blood. Relapse was defined as the re-appearance of hairy cells in the peripheral blood and bone marrow after achieving CR or PR. This was intensively looked for in case of either neutropenia, or thrombocytopenia, or anemia, or all of them were noted again. Toxicity was graded by standard WHO criteria.

Biopsy technique, histological preparations and evaluation

Biopsy cores were taken from the posterior iliac crest with the Jamshidi needle. The cores were employed for histomorphometric measurements. A total of 33 of the biopsy specimens were fixed in Schaffer's fixative, dehydrated, and embedded in plastic (methyl-methacrylate) without decalcification, and sections of 3 m were stained with Gallamin Blue Giemsa for cytology detail and Gomori's stain for reticulin fibers. Histomorphometric measurements were made using a focusing eyepiece ( 8), containing a Zeiss integration disc II. Giemsa stained sections were used for cytologic and quantitative assessment of hairy cells, for which a minimum of 10 microscopic fields were investigated by oil immersion ( 600). For immunohistochemical studies, biopsy specimens were embedded in paraffin and the monoclonal antibodies anti-CD 20 (lineage-specific pan B-cell marker) and DBA. 44 (hairy-cell indicative) were used according to the method previously described¹². Slides were then examined by light microscopy. The percentage of hairy cells per biopsy was quantified by counting 200 CD 20- or DBA. 44-positive cells. All pathologic specimens were reviewed by the same pathologist.

Statistical analysis

Disease-free survival (DFS) was defined for pts achieving CR or PR, and was measured from the first day of treatment with 2-CdA until relapse or death in CR. Observations of DFS were censored at the date of last contact with no report of relapse, and who were known to be alive. Survival (S) was measured both from the day of treatment with 2-CdA as well as from first diagnosis until death from any cause. DFS and S were estimated by the method of Kaplan and Meier.¹³

Results

Patient accrual

A total of 44 consecutive pts were initially treated with a single cycle of 2-CdA. Nine pts, who relapsed after a median of 99 months (61–135), were retreated with a single cycle of 2-CdA, eight of them successfully. All pts were evaluated for clinical and hematological responses. Results were analyzed as of December 31, 2003. The median follow-up from the start of 2-CdA treatment is 8.5 years (0.1–12.2) (Figure 1).

Figure 1.

The curve displays the follow-up of all 44 pts over a maximum time period of 12 years.

Full figure and legend (25K)

In all, 35 (80%) pts complied fully with the planned monitoring, in particular, with regard to the bone marrow aspirates and biopsies within short intervals during the first year following treatment. Compliance during follow-up was reinforced by the attending physician, and two visits in our outpatient clinic scheduled in all pts. Even when bone marrow aspirates and biopsies were refused by the pt, physical examination, full blood work-up and ultrasound of the abdomen were performed in all instances to securely capture secondary malignancy or relapse.

Response

In total, 43 pts (98%) achieved CR (CR-RD n=13), one pt reached a good partial remission of 33 months duration. Those three pts, who were refractory to a preceding treatment attempt using dCF, reached CR following 2-CdA. All 44 pts, including nine pts not complying with the close follow-up by bone marrow aspirate and biopsy, had sustained complete normalization of peripheral blood counts and differentials.

In those pts (n=35) who complied with the very close follow-up by both bone marrow aspirate and core biopsy during the first year and in whom the histological specimens were embedded in plastic (methyl-methacrylate) without decalcification, a CR was diagnosed as judged from the aspirate alone at first evaluation, 2–4 weeks after the end of 2-CdA treatment. In seven of these 35 cases, both bone marrow aspirate and biopsy were simultaneously compatible with a complete disappearance of hairy cells from the bone marrow shortly after treatment. However, in 26/35, the complete clearance of hairy cells from the bone marrow, as evidenced by histology, occurred with a delay of approximately 3–5 months in an average.

Relapse

In all, 17 pts (39%) have relapsed either after CR or PR (one pt) after a median of 48 months (8–131) remission duration. They still showed normal peripheral blood counts at the time of relapse. Five of them were pretreated previously. Pretreatment consisted of either splenectomy followed by IFN, followed by dCF (n=2) or splenectomy only (n=1) or IFN only (n=2). Six pts died in CR of the underlying second malignancy without evidence of residual HCL.

Retreatment at relapse with 2-CdA

So far, nine of the 17 relapsed pts were re-treated with a single cycle of 2-CdA for 7 days. Eight of these reached another CR. One pt did not respond. Myelosuppression following retreatment was not different from that observed during first treatment with 2-CdA. Cytopenias were moderate in the remaining eight relapsed pts with platelet counts not below 10 10⁹/l and leukocytes not below 2.5 G/l, so that retreatment was not yet indicated. Two of the retreated pts were treated a third time with 2-CdA at second relapse: one pt experienced a third CR of 40 months duration, and one pt died immediately after 2-CdA treatment in aplasia due to hemorrhage and septicemia.

Acute toxicities

All pts showed a marked myelosuppression, four requiring G-CSF because of prolonged neutropenia (time to recovery 37–74 days). The median time to nadir for all pts from the start of 2-CdA treatment was 8 days (4–37), for those pretreated for 8.5 days (7–37). The median leukocyte count during nadir was 0.4 10⁹/l (0.1–1.6). The median time to recovery of hemoglobin, platelets and leukocytes for all pts from the end of 2-CdA treatment was 11 days (2–74), for those pretreated 28.5 days (5–63) (Table 2).

Table 2 - Hematological toxicities.

Full table

Extrahematological toxicities

Extrahematological toxicities (WHO grade I–IV) were mainly of grade I and II. In one pt, a grade IV infectious complication occurred, consisting of septic shock with hypotension. However, no causative organism could be identified. The most common toxicities included culture-negative fever (FUO) (36%), headache (19%) and nausea or vomiting (19%). Culture-proven infections were found in four pts (9%) (Table 3). No long-term toxicities with regard to infections or second neoplasias were noted.

Table 3 - Extrahematological Toxicities (%).

Full table

Survival

The median disease-free survival of the 44 study pts from the start of 2-CdA treatment is 8.4 years (0.1–11.4). At 12 years, 36% of pts are still in unmaintained CR (Figure 2). In all, 17 pts have relapsed after a median of 4 years (median 0.7–11). Nine of them received another 7-day cycle of 2-CdA at relapse, two of them a third treatment with 2-CdA at second relapse.

Figure 2.

The curve shows the DFS after the end of initial treatment with 2-CdA until first relapse or death in CR.

Full figure and legend (25K)

The overall survival at 12 years after start of 2-CdA treatment is 79% (Figure 3). So far eight pts died, six of them of an underlying second malignancy, one of a heart attack, and one of hemorrhage and septicemia due to treatment-related aplasia following the third 7-day treatment course at second relapse.

Figure 3.

The curve reflects the overall survival of all 44 pts, eight of them died so far.

Full figure and legend (26K)

The histological subtype of HCL at diagnosis had no impact on reaching a CR, remission duration or survival as observed by others;¹⁰ neither did the type of pretreatment or duration of disease before treatment with 2-CdA.

Discussion

Our study differs from previous similar reports in several aspects: because of a long follow-up, the possibility of obtaining a second and even third response using 2-CdA within the same pt population could be investigated. A very close and consequent follow-up including numerous bone marrow punctures in each pt over a long period of time was a prerequisite for this study. Therefore, the number of pts had to be limited. The frequent bone marrow evaluations, particularly following the first cycle of 2-CdA for induction, by various morphologic and immunologic techniques employed in parallel, revealed new insights as to the dynamics of hairy cell clearing from the bone marrow. Also, the comparison of bone marrow aspiration and histology using core biopsies gave new information as to the value and impact of both techniques in HCL.

The relapse rate, DFS and survival associated with 2-CdA in pts with HCL are not well established, because only a few studies are concerned with a longer follow-up.^{5, 6, 7, 8, 9} This purine analog has dramatically improved the clinical course and prognosis of this disorder, although responsible for neutropenia and lymphocytopenia often associated with fever of unknown origin or proven infectious complications. In our analysis, we wished to address several questions with regard to a long-term follow-up evaluation: what is the long-term prognosis for pts achieving remission with 2-CdA and what is the incidence of relapse? Are there any long-term complications in connection with this kind of treatment? What is the response rate following re-treatment with 2-CdA for pts who relapse?

The decision for evaluating responses as 'complete remission' or 'partial remission' is difficult. In some reports,^{14, 15} no clear distinction and differentiation between aspiration and biopsy findings are undertaken. Surprisingly, this holds true also for the consensus resolution,¹¹ in which the criteria for response are defined. In none of the papers reporting results in HCL, the method of preparation of the bone marrow biopsies is specified, which is important for evaluation, particularly in HCL.¹⁰ Most studies in HCL reporting a high percentage of remissions after treatment with IFN, dCF and 2-CdA are utilizing paraffin for embedding, which is known to cause considerable shrinkage of all types of cells in contrast to the technique using methylmethacrylate.¹⁰ Equally good or perhaps slightly more sensitive than the latter are monoclonal antibodies for the detection of minimal residual infiltration by hairy cells.^{12, 16} These immunological stainings, however, are employed when paraffin is being utilized for embedding. In contrast, they are technically very difficult to perform when methylmethacrylate is used.

It is noteworthy from our study that, when bone marrow aspirate and core biopsy were performed simultaneously (n=35) and the technique of methylmethacrylate for embedding was utilized, 28 pts showed a delayed clearance of hairy cells from the bone marrow of approximately 3–5 months, although in all other parameters, including bone marrow aspiration, CR was diagnosed already 2–4 weeks after the end of 2-CdA treatment. Therefore, in our opinion, histomorphologic and immunohistologic examination should be repeated in those pts at 3 and 5 months, who are not in CR by core biopsy at first evaluation after 2-CdA treatment, but fullfill all other criteria for CR. A protracted effect of this purine analog is not surprising because it exerts its action on resting cells as well, and is not S-phase specific³ Furthermore, the immunological phenotype indicates that hairy cells represent an arrest in late differentiation of B-lymphocytes between those observed in chronic lymphocytic leukemia and myeloma. Also, they are closely related to memory B-cells.¹⁷ Both observations point to a long generation time of hairy cells.

In a recent paper,¹⁸ a definition for complete responders with minimal residual disease (CR-RD) was given. We restricted this to pts with <5% of hairy cells detected by immunocytology and/or immunohistology. In this context, it has been stated¹⁹ that RD detected by monoclonal antibodies may predict relapse in pts with HCL. In our 13 pts in whom CD-RD was diagnosed, three pts relapsed after 28, 61 and 72 months and nine are still in CR after a median of 99 months (61–135). One pt died due to his underlying second malignancy. We could not confirm the observation that RD may identify a subset of pts at risk for relapse. However, the number of pts in our series might be too small.

Due to our stringent criteria of relapse, which were mainly based on the re-appearance of hairy cells in the bone marrow by morphology (5–10%), retreatment of our relapsed pts with 2-CdA was performed only in nine so far when a substantial cytopenia occurred. This time span was approximately 3 months.

The number of relapses in our series is 39%. Due to this low number of relapses and in turn, the long and durable remissions, any prognostic factor analysis such as prior therapy, splenectomy, disease duration, splenic index as performed by others⁵ or the morphological subtype of HCL is difficult to substantiate. For the latter, e.g., it has been described¹⁰ in a large series of HCL pts that the median survival of the ovoid (56 months), the convoluted (12 months) and the indented subtype differed significantly. The impact of none of these factors could be verified in our pts. The reasons for this are, in addition to the low number of relapsed pts, the high CR rate, long duration of DFS and survival. Any of these factors seemed to be outweighted by this extremely effective treatment.

In this context, the risk of second cancer in a large series of pts suffering from HCL has been discussed controversially.^{20, 21, 22} Two reports^{20, 21} suggested an increased incidence of second malignancy, the largest however²² could not support this suspicion. In our small series, eight pts (18%) suffered from additional cancer at start of 2-CdA treatment, six of them died without any evidence of active or residual HCL, two are in CCR. As to a possible role of purine analogs in inducing second malignancies,^{21, 23} no conclusions can be drawn from our study.

dCF, an inhibitor of the enzyme ADA, is another effective drug in HCL^{24, 25, 26} refractory to IFN. Two of our pts who were refractory to dCF reached a CR after administration of 2-CdA. Unlike dCF, 2-CdA is not an ADA inhibitor but rather an ADA-resistant purine substrate analog,³ although both drugs are working within the same biochemical pathway.

Treatment-related toxicity in our pts consisted mainly of myelosuppression, fever of unknown origin, infections, headache and nausea. However, the risk of developing a bacterial infection was limited and confined within 2 weeks after the end of treatment. It has been reported that the administration of 2-CdA induces severe and prolonged lymphopenia with a marked reduction of CD4-positive cells persisting for a long period of time.¹⁴ Despite this immunological impairment, no severe opportunistic infections were seen, neither following second 2-CdA treatment at first relapse nor third treatment with 2-CdA at second relapse. 'Active infection' was an exclusion criterion on the initial NCI trial; however, other large studies with Pentostatin did not exclude pts on this basis.^{24, 25, 26}

Although CdA is highly effective in HCl, even at first relapse, a promising alternative for treating second relapse or primary refractory HCL might be the use of rituximab in the future. First results with this monoclonal antibody in HCL are promising,¹⁸ but also somewhat controversial.²⁷ Rituximab should be explored further in this disease, also with regard to the eradication of minimal disease and in combination with 2-CdA.

References

1. Bouroncle BA, Wiseman BK, Doan CA. Leukemic reticuloendotheliosis. Blood 1958; 13: 609–630. | PubMed | ISI | ChemPort |
2. Golomb HM, Catovsky D, Golde DW. Hairy cell leukemia: a clinical review based on 71 cases. Ann Intern Med 1978; 89: 677–682. | PubMed |
3. Carson DA, Wasson DB, Beutler E. Antileukemic and immunosuppressive activity of 2-chloro-2-deoxy-adenosine. Proc Natl Acad Sci USA 1984; 81: 2232–2236. | PubMed |
4. Piro LD, Carrera CJ, Carson DA, Beutler E. Lasting remissions in hairy cell leukemia induced by a single infusion of 2-chlorodeoxyadenosine. New Engl J Med 1990; 322: 1117–1121. | PubMed |
5. Tallman MS, Hakimian D, Rademaker AW, Zanzig C, Wollins E, Rose E et al. Relapse of hairy cell leukemia after 2-chlorodeoxyadenosine: long-term follow-up of the Northwestern University experience. Blood 1996; 88: 1954–1959. | PubMed |
6. Hoffman MA, Janson D, Rose E, Rai KR. Treatment of hairy-cell leukemia with Cladribine: response, toxicity, and long-term follow-up. JCO 1997; 15: 1138–1142.
7. Lauria F, Rondelli D, Zinzani PL, Bocchia M, Marotta G, Salvucci M et al. Remission in patients with hairy cell leukemia treated with 2-CdA: a 5-year survey. Leukemia 1997; 11: 629–632. | Article | PubMed |
8. Saven A, Burian C, Koziol JA, Piro LD. Long-term follow-up of patients with hairy cell leukemia after Cladribine treatment. Blood 1998; 92: 1918–1926. | PubMed |
9. Jehn U, Bartl R, Dietzfelbinger H, Vehling-Kaiser U, Wolf-Hornung B, Hill W et al. Long-term outcome of hairy cell leukemia (HCL) treated with 2-chlorodeoxyadenosine (2-CdA). Ann Hematol 1999; 78: 139–144. | Article | PubMed |
10. Bartl R, Frisch B, Hill W, Burkhardt R, Sommerfeld W, Sund M. Bone marrow histology in hairy cell leukemia: identification of subtypes and their prognostic significance. Am J Clin Pathol 1983; 79: 531–545. | PubMed |
11. Anonymous. Consensus resolution. Proposed criteria for evaluation of response to treatment inhairy cell leukemia. 1987; 1: 405–408.
12. Hakimian D, Tallman MS, Kiley C, Peterson LAC. Detection of minimal residual disease by immunostaining of bone marrow biopsies after 2-chorodeoxyadenosine for hairy cell leukemia. Blood 1993; 82: 1798–1802. | PubMed |
13. Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. JAMA 1958; 53: 457–481.
14. Estey EH, Kurzrock R, Kantarjian HM, O'Brien SM, McCredie KB, Beran M et al. Treatment of hairy cell leukemia with 2-chloro-deoxyadenosine (2-CdA). Blood 1992; 79: 882–887. | PubMed |
15. Tallman MS, Hakimian D, Variakojis D, Koslow D, Sisney GA, Rademaker AW et al. A single cycle of 2-chlorodeoxyadenosine results in complete remission in the majority of patients with hairy cell leukemia. Blood 1992; 80: 2203–2209. | PubMed |
16. Thaler J, Dietze O, Faber V, Greil R, Denz H, Ho AD et al. Monoclonal antibody B-ly 7: a sensitive marker for detection of minimal residual disease in hairy cell leukemia. Leukemia 1990; 4: 170–175. | PubMed |
17. Burns GF, Cawley JC, Worman CP. Multiple heavy chain isotypes on the surface of the cells of hairy cell leukemia. Blood 1978; 52: 1132–1136. | PubMed |
18. Thomas DA, O'Brien S, Bueso-Ramos C, Faderl ST, Keating MJ, Giles FJ et al. Rituximab in relapsed or refractory hairy cell leukemia. Blood 2003; 102: 3906–3911. | Article | PubMed |
19. Wheaton S, Tallman MS, Hakimian D, Peterson L. Minimal residual disease may predict bone marrow relapse in patients with hairy cell leukemia treated with 2-chlorodeoxyadenosine. Blood 1996; 87: 1556–1560. | PubMed |
20. Kurzrock R, Strom SS, Estey E, O'Brien S, Keating MJ, Jiang H et al. Second cancer risk in hairy cell leukemia: analysis of 350 patients. JCO 1997; 15: 1803–1810.
21. Cheson BD, Vena DA, Barett J, Freidlin B. Second malignancies as a consequence of nucleoside analog therapy for chronic lymphoid leukemias. JCO 1999; 17: 2454–2460.
22. Federico M, Zinzani PL, Frassoldati A, Vinceti M, Mode A, Annino L et al. Risk of second cancer in patients with hairy cell leukemia: long term follow-up. JCO 2002; 20: 638–646. | Article |
23. Schlaifer D, Rigal-Huguet F, Pris J. Secondary neoplasms in two patients treated with purine analogues. Nouo Rev Fr Hematol 1994; 36: 321 (letter).
24. Ho AD, Thaler J, Mandelli F, Lauria F, Zittoun R, Willemze R, et al., for the European Organization for the Research and Treatment of Cancer Leukemia Cooperative Group. Response to pentostatin in hairy-cell-leukemia refractory to interferon alpha. J Clin Oncol 1989; 7: 1533–1538. | PubMed |
25. Kraut EH, Bouroncle BA, Grever MR. Low-dose deoxycoformycin in the treatment of hairy cell leukemia. Blood 1986; 68: 1119–1122. | PubMed |
26. Spiers ADS, Moore D, Cassileth PA, Harrington D, Cummings F, Neiman R et al. Remissions in hairy-cell-leukemia with pentostatin. N Engl J Med 1987; 316: 825–830. | PubMed |
27. Nieva J, Bethel K, Saven A. Phase 2 study of rituximab in the treatment of cladribin-failed patients with hairy cell leukemia. Blood 2003; 102: 810–813. | Article | PubMed |