|
|
| |||||||||||||||||||||||||||||||||||||||
Background Juvenile chronic myelogenous leukemia (CML) is a rare myeloproliferative disease of infants and young children for which there is no effective therapy other than allogeneic bone marrow transplantation. In vitro, isotretinoin (13-cis-retinoic acid) attenuates both the spontaneous proliferation of leukemic peripheral-blood progenitor cells (granulocyte-macrophage colony-forming units) and their selective hypersensitivity to granulocyte-macrophage colony-stimulating factor (GM-CSF). We conducted a pilot study to evaluate the clinical efficacy of isotretinoin in juvenile CML.
Methods To be eligible the patients had to have newly diagnosed untreated disease, leukocytosis with monocytosis, marrow with less than 25 percent blasts, hepatosplenomegaly, no chromosomal abnormalities, and negative viral cultures and antibody titers. Isotretinoin was administered orally in single daily doses of 100 mg per square meter of body-surface area. When possible, patients subsequently underwent bone marrow transplantation.
Results Ten children (median age, 10 months) were enrolled in the study. In all 10 there was spontaneous colony formation of leukemic progenitor cells in vitro. In the eight patients tested there was hypersensitivity to GM-CSF. The only toxic effect of isotretinoin therapy was cheilitis in two patients. Four children had disease progression. Two children had complete responses to isotretinoin (normalization of the white-cell count and disappearance of organomegaly), three had partial responses (more than a 50 percent reduction in the white-cell count and degree of organomegaly), and one had a minimal response (more than a 50 percent reduction in the white-cell count, but a 26 to 50 percent reduction in the degree of organomegaly). The median duration of response was 37 months (range, 6 to 83). Three of the four children who had a complete or partial response and who did not undergo bone marrow transplantation were alive 36 to 83 months after the diagnosis of juvenile CML. The spontaneous colony formation in vitro was reduced in samples from the five patients in whom this factor was reassessed during treatment. There was also a reduction in the hypersensitivity of leukemic progenitor cells to GM-CSF in the two patients retested.
Conclusions Isotretinoin can induce durable clinical and laboratory responses in patients with juvenile CML.
9. Various chemotherapy regimens have been used in patients with juvenile CML, with well-documented responses but a limited effect on outcome.10,11,12 Only allogeneic bone marrow transplantation has resulted in long-term control of this disease13 although relapse remains an important problem.
Isotretinoin (13-cis-retinoic acid), an analogue of vitamin A, is known to inhibit carcinogenesis14,15 in vitro and in vivo and to induce terminal granulocytic differentiation in selected lines of myeloid leukemia cells and in cultures of samples from some patients with leukemia.16,17,18,19,20 Isotretinoin may be effective in myelodysplastic syndromes of adults,21 although its long-term benefits have been questioned22.
In an in vitro study of peripheral-blood samples from four children with juvenile CML, we found a 60 to 100 percent rate of inhibition of the spontaneous proliferation of CFU-GM in the presence of isotretinoin at a concentration of 0.1 to 1 micro M,23 concentrations achievable with oral administration of the drug and with acceptable toxicity. These in vitro data have been confirmed in 52 patients with previously treated or untreated juvenile CML (data not shown).
A prospective, multi-institutional pilot study was conducted to explore the clinical efficacy and toxicity of oral isotretinoin in children with previously untreated juvenile CML and to determine the relation, if any, between changes in peripheral-blood CFU-GM in culture and the clinical response to treatment with isotretinoin. The preliminary results of this trial have been presented previously24.
Methods
All patients with newly diagnosed untreated juvenile CML were eligible for the study. Before enrollment, informed consent was obtained from the patients or their parents according to Food and Drug Administration and institutional guidelines. The diagnostic criteria for juvenile CML used in this study were a compilation of previously reported clinical and laboratory features3,4,5,6 and included leukocytosis with monocytosis (values >2 SD above the mean for age), the presence of immature myeloid precursors in the peripheral blood, less than 25 percent blasts in marrow, the absence of chromosomal abnormalities, the presence of hepatosplenomegaly, negative viral antibody titers and cultures, and spontaneous proliferation of peripheral-blood CFU-GM in cultures. If sufficient cells were available, selective hypersensitivity of CFU-GM to GM-CSF was also studied. The methods used for the in vitro studies have been previously reported.6,7,8 All patients were required to have normal liver and renal function. If clinically stable, the study patients were observed for spontaneous improvement for up to one month without treatment. No such remissions were observed.
Eligible patients received isotretinoin in a single daily dose of 100 mg per square meter of body-surface area orally. The doses were individually prepared by puncturing a capsule of isotretinoin (Accutane, Roche Dermatologics, Hoffmann-La Roche), withdrawing the required amount, and mixing it with an equal volume of safflower oil to reduce mucous-membrane contact with the drug.
The patients were monitored periodically by means of physical examinations, complete blood counts, and blood chemical analyses. Isotretinoin concentrations were not measured consistently. Therapy was continued as long as a clinical response was observed or the patient's condition remained stable or until the patient underwent bone marrow transplantation. The dose of isotretinoin could be increased by 25 percent each week if, after one month of therapy, there was minimal response and no toxic effects.
Responses to isotretinoin therapy were defined as follows: complete response, normalization of the white-cell count and disappearance of organomegaly; partial response, more than a 50 percent reduction in the white-cell count and the degree of organomegaly; minimal response, a reduction of 26 to 50 percent in the white-cell count and the degree of organomegaly, or a reduction in the white-cell count of more than 50 percent but no change in the degree of organomegaly, or a reduction of more than 50 percent in the degree of organomegaly but no change in the white-cell count; and no response, a reduction in the white-cell count and degree of organomegaly of 25 percent or less. The disease was considered to have progressed if the white-cell count or degree of organomegaly increased by more than 25 percent.
Since many patients were receiving transfusions of packed cells or platelets, neither hemoglobin concentrations nor platelet counts were used to monitor response. When patients no longer required blood products their disease was considered stable. The classic definition of minimal response was expanded to account for patients who exceeded its criteria but met only one criterion for a partial response.
Results
From September 1985 to June 1991 we enrolled 14 patients who fulfilled the diagnostic criteria for juvenile CML in the study. Four patients could not be evaluated for a response or toxic effects because of early death in one patient and inadequate data in three. The clinical characteristics of the remaining 10 patients are listed in Table 1.
|
All patients received at least 100 mg of isotretinoin per square meter, and in some the dose was increased to as much as 200 mg per square meter. The only toxic effect encountered was a moderate degree of cheilitis in two patients. This was not severe enough to warrant discontinuation of the drug, but the dose was temporarily decreased in one patient. No hepatotoxicity, hypertriglyceridemia, or growth retardation was observed, even in patients receiving isotretinoin for an extended period.
The patients' clinical responses are listed in Table 2. Two patients had complete responses, three had partial responses, and one had a minimal response. Subjectively, patients were noted to look and feel better after isotretinoin therapy was begun, even before an objective clinical effect was documented. Four patients had progressive disease (Table 2). One of these patients (Patient 7) had a 40 percent increase in the white-cell count with a 34 percent decrease in the size of the liver and spleen. In one (Patient 9) the disease progressed quickly over a two-week period, with a rapid increase in both the white-cell count and the degree of organomegaly. In the other two patients (Patients 8 and 10) the disease progressed over a period of one to two months.
|
The patients' outcomes are listed in Table 3. Patient 1 discontinued isotretinoin therapy after 1 year of continuous complete remission and was still free of disease at the most recent follow-up 83 months later. Patient 2 was treated the longest with isotretinoin, because there was no appropriate donor for bone marrow transplantation. His disease was stable for approximately 3 1/2 years during treatment with isotretinoin before progressing, and he eventually died of juvenile CML. Patient 3 had a very good partial response (reduction in white-cell count to <20,000 per cubic millimeter and >50 percent reduction in the size of the liver and spleen over a 6-month period) that at the time of this writing had been maintained for 36 months. Because of the child's overall good health, his parents have refused to allow him to undergo allogeneic bone marrow transplantation from an HLA-matched, unrelated donor. After 9 months of isotretinoin therapy, Patient 4 had what was considered to be a partial response (a complete response in terms of the white-cell count but a minimal response in terms of the degree of organomegaly) and as of the most recent follow-up 37 months later was continuing to receive isotretinoin and had stable disease. Thus far, the linear growth rates of Patients 2, 3, and 4 have been normal during extended isotretinoin therapy. All three patients who underwent bone marrow transplantation (Patients 5, 6, and 7) were alive and free of disease 21 to 42 months later; Patient 5 has growth retardation, and Patient 7 has chronic graft-versus-host disease.
|
|
|
We document durable, objective responses to orally administered isotretinoin in patients with previously untreated juvenile CML, with minimal toxic effects. Six of the 10 young children studied had complete, partial, or minimal responses lasting 6 to 83 months. At the most recent follow-up, three patients, two of whom were still taking isotretinoin, remained in stable condition. This ability of isotretinoin to control juvenile CML for extended periods, thus allowing time to ready the patient for bone marrow transplantation, is at least comparable to that of other types of chemotherapy, although we have treated only a limited number of patients. Two of the three patients described by Lilleyman et al.10 who were treated with low-dose cytarabine and oral mercaptopurine had complete responses and survived 9 to 26 months. There was one long-term survivor. In a retrospective review, Chan and colleagues12 described complete responses and survival of 21 to more than 32 months in four patients with juvenile CML who received myelosuppressive therapy similar to that used for acute myeloid leukemia; however, neither of two children who received single-agent therapy had a complete response or survived. An advantage of dose-intensive chemotherapy is the rapidity of the response, although this has not translated into longer survival for patients who do not undergo bone marrow transplantation and has resulted in substantial toxic effects. In contrast, in our study orally administered isotretinoin prolonged survival beyond the anticipated mean of less than nine months.1 There were no serious untoward effects, and three of the four children with objective responses who did not undergo bone marrow transplantation survived for long periods in relatively good health.
The responses to isotretinoin appear to be biphasic and relatively slow. During the initial two weeks of therapy there is a demonstrable decline in the white-cell count, followed by a maximal decline in the white-cell count and degree of organomegaly over a median period of two months. The initial phase may be related to either a direct toxic effect of isotretinoin on leukemic progenitor cells or the induction of apoptosis, as has been observed in culture models of HL-60 cells.25 The secondary phase of the response probably represents a direct effect involving either the induction of differentiation16,17,18 or modulation of hypersensitivity to GM-CSF, as observed in Patients 5 and 6. The sluggish response may compromise the efficacy of isotretinoin in patients such as Patient 9, who had a relatively rapid progression of disease. This disadvantage might be obviated in future trials by combining low-dose cytarabine with isotretinoin for induction therapy and using isotretinoin for continuation therapy.
The precise biologic mechanism of the effect of isotretinoin in juvenile CML is still under investigation. We have previously reported that the spontaneous proliferation in vitro of peripheral-blood progenitors from eight patients with juvenile CML was regulated by GM-CSF on the basis of the selective hypersensitivity to GM-CSF of the leukemic progenitor cells7,8. All of the 34 previously treated and untreated patients studied, including 8 described in this report, have had this feature. These observations raise the prospect that selective hypersensitivity to GM-CSF is a pathognomonic marker for juvenile CML. The fact that peripheral-blood samples from both patients in this series whose hypersensitivity to GM-CSF was reevaluated during treatment with isotretinoin showed an improvement in the dose-response curve, as compared with pretreatment samples, suggests that there is some relation between the administration of isotretinoin and the attenuation or modulation of hypersensitivity.
In this limited series, there were no identifiable patterns of clinical or laboratory features, including the presence of neurofibromatosis, that predicted a response to isotretinoin. If a biologic effect of the drug is the principal basis for disease control in juvenile CML, one would anticipate some response in all patients. That this was not the case suggests two possible explanations. First, the bioavailability of isotretinoin may have been variable because of poor preparation or other pharmacologic factors. Since serum concentrations of isotretinoin were not measured consistently in this pilot study, such a possibility cannot be assessed. Second, although GM-CFU from all patients with juvenile CML appear to have selective hypersensitivity to GM-CSF in vitro, other secondary or indirect endogenous regulators such as tumor necrosis factor a9 may have a role in the response to isotretinoin or in the overall clinical course of juvenile CML, which has substantial heterogeneity1,2,3.
In conclusion, treatment with isotretinoin alone induces durable clinical and laboratory responses in patients with juvenile CML and warrants further evaluation in a phase 2 study of previously untreated patients to establish more accurately the rate of response. Such an investigation is currently being conducted by the Pediatric Oncology Group. Although laboratory investigations have provided some understanding of the pathogenesis of myeloproliferation in juvenile CML, ongoing efforts to delineate the cellular biologic properties related to hypersensitivity to GM-CSF and the metabolism and effect of retinoids will probably yield further insight useful in developing other treatments for this devastating disease.
Supported in part by grants from the National Cancer Institute (CA-25408, CA-29139, CA-11233, and CA-56072), the Children's Cancer Research Fund, and Cancer Research and Youth Outreach Network, Inc. Dr. Emanuel is a Special Fellow of the Leukemia Society of America.
We are indebted to Mr. Gary Martin, Ms. Lana Bates-Atchely, and Ms. Jitka Sokol for excellent technical support, to Mr. Lance Lang for assistance in data management, and to Ms. Cathy Anderson and Ms. Anita Mollica for assistance in the preparation of the manuscript.
Source Information
From the University of Alabama at Birmingham, Birmingham (R.P.C., P.D.E., R.J.G.); H.L. Moffitt Cancer Center, Tampa, Fla. (K.S.Z.); Northwestern University, Chicago (S. Cohn, L.S.); Eastern Virginia Medical School, Norfolk (R.L.B.); University of Vermont School of Medicine, Burlington (A.H.); Dartmouth-Hitchcock Medical Center, Lebanon, N.H. (S. Chaffee); and the University of Oklahoma, Oklahoma City (R.N.).
Address reprint requests to Dr. Castleberry at the Division of Pediatric Hematology/Oncology, Children's Hospital, 1600 7th Ave. S., Birmingham, AL 35233.
References
| |||||||||||||||||||||||||||||||||||||||
Related Letters:
Isotretinoin for Juvenile Chronic Myelogenous Leukemia
Pui C.-H., Aricò M., Donadieu J., Castleberry R. P., Emanuel P. D.
Extract |
Full Text
N Engl J Med 1995;
332:1520-1521, Jun 1, 1995.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
