Since the development of the French–American–British (FAB) classification, chronic myelomonocytic leukaemia (CMML) has been listed as one of the myelodysplastic syndromes (MDS) ( Bennett et al, 1982 ). However, the disease has many features of a myeloproliferative disorder (MPD) ( Bain, 1999) and the World Health Organization (WHO) has recently recategorized CMML as an MPD. Distinctions from other MPD have been described ( Michaux & Martiat, 1993; Bennett et al, 1994 ; Cambier et al, 1997 ). Regardless of categorization, however, and despite the ‘chronic’ course suggested by the terminology, CMML is a progressive disease with a high rate of transformation into acute leukaemia ( Tefferi et al, 1989; Greenberg et al, 1997 ). Many patients will respond to chemotherapy, but complete responses are infrequent and remission duration is generally short. Wattel et al (1996) reported a randomized trial comparing hydroxyurea with etoposide in 105 patients with CMML. Responses to treatment were observed in 60% and 36% of patients, respectively, but median survivals were only 20 months and 9 months for the two treatment arms respectively. Similarly, other trials using low-dose cytosine arabinoside, intensive chemotherapy, topotecan or all-trans retinoic acid showed only palliative effects ( Cheson et al, 1986 ; Fenaux et al, 1991 ; Beran et al, 1996 ; Cambier et al, 1996 ). Considering those results and in view of our current understanding that CMML is a disease of haemopoietic stem cells, treatment of CMML with haemopoietic stem cell transplantation is an attractive alternative. Here, we report our experience with 21 patients with CMML who underwent allogeneic marrow transplantation at the Fred Hutchinson Cancer Research Center (FHCRC).
We evaluated the outcome of allogeneic bone marrow transplantation (BMT) in 21 patients with chronic myelomonocytic leukaemia (CMML) who were treated at the Fred Hutchinson Cancer Research Center between 1990 and 1998. There were 11 male and 10 female patients with a median age of 47·4 years (range 1·0–62·9). Patients were conditioned either with total body irradiation (TBI) and chemotherapy, with or without antithymocyte globulin (n = 19), or with chemotherapy alone (n = 2). The marrow donor was an HLA-identical sibling in 12 patients, an HLA-non-identical related donor in three patients and an unrelated volunteer donor in six patients. All evaluable patients achieved sustained engraftment. Fifteen patients developed grades II–IV acute graft-versus-host disease (GVHD). Nine patients (43·0%) are surviving disease free at 0·7–8·1 years (median 6·9) after transplantation. Five patients relapsed 75–660 d after transplant and all died. Five patients died with organ failure and two died with GVHD and associated infections. The Kaplan–Meier estimates of disease-free survival and relapse at 3 years were 39% and 25% respectively. The probability of survival was improved in patients with shorter disease duration compared with those with a long interval from diagnosis to BMT. Thus, as with other myeloproliferative diseases or myelodysplastic syndromes, BMT offers curative therapy for a proportion of patients with CMML. We suggest that patients with CMML who have a suitable donor should be considered for transplantation, probably early in their disease course. However, it will be important to develop new regimens with enhanced antileukaemic efficacy without further increasing regimen-related toxicity and mortality.
PATIENTS and METHODS
Patients. Twenty-one patients with CMML as defined by FAB criteria ( Bennett et al, 1982 ) were transplanted at the FHCRC between June 1990 and July 1998. Patient and transplant characteristics are summarized in Tables I and II. Data on some of these patients were included in previous publications ( Anderson et al, 1996 ; Appelbaum & Anderson, 1998). Eleven patients were men and 10 were women, and the median age at the time of transplant was 47 years. The median Karnofsky performance status (KPS) at the time of transplant was 80%. The median time from diagnosis to transplantation was 9 months. All 21 patients had de novo CMML. At the time of transplantation, 12 patients had a white blood cell count (WBC) of > 12 × 109/l (proliferative CMML) and nine patients had a WBC of ≤ 12 × 109/l (non-proliferative CMML) ( Greenberg et al, 1997 ). Two patients received intensive chemotherapy before transplantation because of progression to acute myelogenous leukaemia in one patient and a blast count of 9% in the second patient. Seven patients received no treatment or only transfusion support, 10 patients received hydroxyurea and two patients received prednisolone or erythropoietin. Three patients had undergone splenectomy. In nine patients, marrow examination revealed < 5% blasts at the time of transplant. Twelve patients had excess blasts, ranging from 5% to 25%; three of these were categorized as proliferative CMML. Twelve patients did not show cytogenetic abnormalities at the time of transplantation, monosomy 7 was present in three patients and other cytogenetic abnormalities were observed in the remaining six patients. As proposed by the international prognostic scoring system (IPSS) ( Greenberg et al, 1997 ), only the nine patients with non-proliferative CMML were scored. Among these, one patient qualified as low risk, five as intermediate 1 and two as intermediate 2; in one patient, the data required for categorization were incomplete.
Transplantation procedure. Transplant characteristics are summarized in Table II. Typing of patients and donors for HLA-A, -B, -C, -DRB1 and -DQB1 was performed as previously described ( Petersdorf et al, 1998 ). Twelve patients (57%) received marrow from an HLA-matched sibling, three from an HLA-haploidentical sibling or son incompatible for one (n = 1) or multiple (n = 2) loci, four from an HLA-matched unrelated donor and two from an unrelated donor incompatible for one patient at DQB1 and for the other patient at DRB1 plus DQB1. The preparative regimens used to treat patients were selected based on concurrently ongoing studies involving patients with myelodysplasia. Two patients were conditioned with cyclophosphamide (CY) 120 mg/kg and fractionated total body irradiation (TBI), with a dose of 14·4 Gy for one child and 13·2 Gy for one adult patient. Two patients received busulphan (BU) 16 mg/kg and CY 120 mg/kg, nine patients (43%) received BU 7 mg/kg, CY 50 mg/kg and 12 Gy fractionated TBI, and the remaining seven patients received BU (7 mg/kg) and 12 Gy fractionated TBI ( Anderson et al, 1996 ; Appelbaum & Anderson, 1998). The median nucleated marrow cell dose administered was 2·2 × 108/kg (range 0·37–4·72 × 108/kg). All patients received unmodified marrow. Graft-versus-host disease (GVHD) prophylaxis included cyclosporin (CSP) plus methotrexate (MTX) in 17 patients, CSP plus prednisolone in two patients and FK-506 plus MTX in two patients ( Storb et al, 1986; Nash et al, 1996; Deeg et al, 1997 ).
Definition of end-points. Engraftment was defined as achieving a sustained granulocyte level of ≥ 0·5 × 109/l and surviving for at least 21 d. GVHD was assessed by established criteria ( Storb et al, 1986 ). Survival was the time from transplant until death or date of last contact. Disease-free survival (DFS) was the time from transplant until relapse or death due to a non-relapse cause. Relapse was defined as the reappearance of morphological criteria of CMML or recurrence of previously present cytogenetic abnormalities.
Statistical considerations. Actuarial DFS was calculated using the Kaplan–Meier method ( Kaplan & Meier, 1958). The probability of relapse was expressed as cumulative incidence. The hazard ratio of the impact of disease duration on outcome was determined by Cox regression ( Cox, 1972) and the hazard of mortality dependent upon disease duration was compared by log-rank test ( Kalbfleisch & Prentice, 1980). Data were analysed as of 1 July 1999.
The results are summarized in Table III.
One patient died on day 18 and was not evaluable for engraftment. The remaining 20 patients achieved sustained engraftment. A neutrophil count of 0·5 × 109/l was reached at a median of 20 d (range 11–30); a platelet count of 20 × 109/l was reached in 17 patients at a median of 22 d (range 18–58), whereas three patients died before reaching that level.
Six patients with documented engraftment showed no (grade 0) or minimal (grade I) acute GVHD. Fourteen patients with engraftment and the patient who died on day 18 developed grades II–IV acute GVHD. Chronic GVHD developed in 14 of the 16 patients at risk (surviving beyond 100 d); it was limited in nine and extensive in five patients. At latest follow-up, two of these patients still required immunosuppressive therapy for chronic GVHD.
Five patients experienced a recurrence of CMML at 79–657 d (median 161) after transplantation. The cumulative incidence of relapse at 3 years was 25%. All but one of these patients had shown excess blasts at the time of transplant.
Currently, nine patients (43%) are surviving 0·7–8·1 years (median 6·9) after transplantation ( Fig 1). Six of these patients had excess blasts at transplantation; three had non-proliferative and six had proliferative CMML. Specifically, three out of 12 patients transplanted from an HLA-identical sibling, three out of three patients transplanted from an HLA-non-identical relative and three out of six patients transplanted from an unrelated donor are surviving without evidence of disease. Patients with a shorter disease duration appeared to fare better, with six out of 11 patients (54%) surviving with a disease duration of less than 12 months compared with three out of 10 patients (30%) with a disease duration of 12 months or longer (P = 0·05). The hazard ratio for these two groups was 3·09 [95% confidence interval (CI) 0·95, 10·06].
The survival rate was similar for different conditioning regimens, with three out of nine patients surviving after BU/CY/TBI conditioning, three out of seven patients surviving after BU/TBI conditioning, two out of three patients surviving after CY/TBI [± antithymocyte globulin (ATG)] conditioning and one out of two patients surviving after BU/CY conditioning. At the time of last contact, all of these disease-free survivors had a KPS of more than 90% (100% in five patients).
Causes of death
As shown in Table IV, five patients died with organ failure, two with chronic GVHD and viral infections and five patients died with a relapse of CMML.
CMML is a haemopoietic stem cell disorder that has been considered to be among the diagnoses listed under myelodysplastic syndrome as well as among myeloproliferative disorders ( Bennett et al, 1982 ; Bain, 1999). CMML is characterized by the prominence of monocytes in marrow and peripheral blood. Peripheral blood cytopenias, if present, may not be severe. The course is often protracted; however, the tempo of the disease tends to accelerate over time with conversion into acute myeloid leukaemia ( Fenaux et al, 1987 , 1988; Worsley et al, 1988 ). The marked proliferative tendency and the frequent paucity of dysplastic features were causes for a recent reclassification of CMML as a myeloproliferative, rather than dysplastic, disorder ( Vallespi et al, 1998 ).
Past experience shows that complete remissions are difficult to achieve with conventional chemotherapy and are generally of short duration ( Fenaux et al, 1991 ; Beran et al, 1996 ). The high success rate with marrow transplantation in patients with chronic myelogenous leukaemia suggests that haemopoietic stem cell transplantation offers potentially curative therapy for patients with other MPDs ( Gratwohl et al, 1993 ; Appelbaum et al, 1995 ). In keeping with this concept, high-dose chemoradiotherapy followed by allogeneic marrow or peripheral blood stem cell transplantation may provide an attractive approach to the treatment of patients with CMML. The data presented here support this concept. Considering that the median age at transplant was almost 50 years, the results are encouraging, with a Kaplan–Meier estimate of relapse-free survival at 3 years of 39%.
Although the number of patients included in this analysis was too small for a comprehensive statistical analysis, several findings were of possible significance. The interval from disease diagnosis to transplantation appeared to have an impact on post-transplant outcome: the median disease duration in surviving patients was 6 months compared with 16 months in patients who died after transplant. The hazard of dying after transplant among patients with a disease duration of more than 1 year was three times higher than among patients who were transplanted within 1 year of diagnosis. A similar relationship between disease duration and transplant outcome has been observed for other disorders, in particular for patients with chronic myelogenous leukaemia ( Bacigalupo et al, 1993; Goldman et al, 1993 ; Thomas & Clift, 1999), although it is not clear why disease duration has a negative effect. Conceivably, an ‘acceleration’ of the disease with an increase in the proportion of blasts will increase the risk of failure. In fact, although only one post-transplant relapse occurred among nine patients with less than 5% blasts in the marrow, four out of 12 patients with excess blasts experienced disease recurrence. In addition, both patients who had received combination chemotherapy for remission induction before transplant died with multiorgan failure thought to be a result of regimen-related toxicity. On the other hand, patients with clonal cytogenetic abnormalities fared as well as patients with a normal karyotype and survival among patients with or without excess blasts was similar. Overall, the results show that patients with CMML who have a suitably matched related or unrelated stem cell donor should be considered for transplantation, probably early in their disease course.
We thank E. Bryant Ph.D. for providing cytogenetic data and B. Larson and H. Crawford for typing the manuscript. This study was supported in part by PHS grants CA18029, CA15704 and HL36444 awarded by the National Institutes of Health, DHHS, Bethesda, MD, USA; H.J.D. is also supported by a grant from the Gabriella Rich Foundation and J.E.A. by an American Cancer Society Clinical Oncology Career Development Award.