- © 2002 by American Society of Clinical Oncology
Intensified and Shortened Cyclical Chemotherapy for Adult Acute Lymphoblastic Leukemia
- Address reprint requests to Charles A. Linker, MD, Bone Marrow Transplant Program, University of California Medical Center, 400 Parnassus Ave, Rm A502, San Francisco, CA 94143-0324; email: linkerc{at}medicine.ucsf.edu
Abstract
PURPOSE: To assess the efficacy and toxicity of a new treatment program of intensified and shortened cyclical chemotherapy (protocol 8707) in adults with acute lymphoblastic leukemia (ALL).
PATIENTS AND METHODS: Previously untreated adults ≤ 60 years old with ALL were treated with a four-agent induction chemotherapy regimen. This was followed by cyclical postremission therapy with high-dose cytarabine/etoposide; high-dose methotrexate/6-mercaptopurine; and daunorubicin, vincristine, prednisone, and asparaginase. Maintenance chemotherapy with oral methotrexate and 6-mercaptopurine was continued for 30 months. CNS prophylaxis was given with intrathecal methotrexate in addition to the systemic chemotherapy indicated above.
RESULTS: Seventy-eight of 84 patients (93%) achieved complete remission. With a median follow-up of 5.6 years, 5-year event-free survival (EFS) of all remission patients is 52%. Patients with high-risk features including adverse cytogenetics, failure to achieve remission with the first cycle of chemotherapy, and B-precursor disease with WBC counts more than 100,000/μL all relapsed unless taken off study for transplantation. For patients without these high-risk features, 5-year EFS was 60%. Compared with our previous treatment regimen, results appear to be improved for patients with standard-risk B-precursor disease (5-year EFS, 66% v 34%; P = .01).
CONCLUSION: Intensified and shortened chemotherapy may improve the outcome for patients with ALL with B-precursor disease lacking high-risk features. Further trials of this regimen are warranted.
DURING THE 1980s, significant progress was made in the treatment of adult lymphoblastic leukemia (ALL).1 In both single-institution and cooperative group studies, intensive multiagent chemotherapy produced complete remission (CR) rates of 80% to 90% in adults younger than age 60.2,3 A variety of treatment programs that included intensified postremission chemotherapy produced overall 5-year disease-free survival (DFS) rates of 35% to 40%.2-5 We learned to identify prognostic factors that were important in predicting outcome.1-6 These included age, WBC count, phenotype, time to achieve CR, and cytogenetics. With the more routine use of cytogenetic testing and with molecular probes that could identify bcr-abl and MLL gene abnormalities, the role of cytogenetics in predicting outcome has gained increasing importance.7-9
During the past 10 years, progress in the treatment of adult ALL has proceeded at a slower pace. We previously reported on the use of a treatment regimen (protocol 8001) using four-agent induction chemotherapy followed by cyclical postremission chemotherapy.2 We designed the current treatment program (protocol 8707) with the goals of improving outcomes and decreasing certain side effects and toxicities.
To decrease the toxicities associated with cranial irradiation used as part of CNS prophylaxis, we omitted cranial irradiation in patients lacking overt CNS disease. We continued the use of intrathecal methotrexate and used repeated doses of systemic chemotherapy (cytarabine, methotrexate, and etoposide) in doses expected to give good CNS penetration. To decrease the incidence of cardiomyopathy seen in our previous study, we lowered the cumulative dose of daunorubicin from 600 mg/m2 to 360 to 420 mg/m2. This was done using two treatment modules in which the dose of daunorubicin was at least 180 mg/m2 so that the dose intensity of anthracycline therapy would not be compromised.
Although patients with T-cell ALL fared well in our previous study with long-term DFS of 55%, patients with B-precursor disease fared much less well, with DFS of 27%.2 Patients with high-risk features such as t(9,22) and t(4,11) contributed to this overall outcome of the B-precursor group. However, patients with B-precursor disease lacking these high-risk features had a 5-year DFS of only 34%. A specific goal of the new treatment program was to improve outcomes for these patients. In order to address this population, high-dose methotrexate was used much earlier in this treatment program and in repeated doses. Other changes were also made including increasing the dose of cytarabine and podophyllotoxin.
We thought it unlikely that this intensified chemotherapy program would be able to overcome the dismal prognosis of patients with high-risk features such as adverse cytogenetics and failure to achieve remission with a first treatment course. In order to improve outcome in these patients, autologous stem-cell transplantation was added as a treatment option for those who were not candidates for allogeneic transplantation (which was our treatment of choice for such patients).
PATIENTS AND METHODS
Eighty-four consecutive adults aged 16 to 59 were enrolled either at the University of California, San Francisco (85% of patients), or were treated by collaborating physicians in the community. Patients were enrolled between September 1987 and January 1998. All patients gave written informed consent according to guidelines at each institution’s committee on human research.
Patients were previously untreated and had either ALL or high-risk lymphoblastic lymphoma. Patients with mature B-cell ALL (Burkitt’s ALL) were excluded. The diagnosis of ALL was made according to French-American-British criteria. Morphology had to be compatible with ALL and histochemistry negative for peroxidase and alpha-naphthyl butyrate esterase. In addition, we required an immunophenotype compatible with ALL. Before June 1989, four of 18 patients were not immunophenotyped and were classified as ALL on the basis of the strong expression of terminal deoxynucleotidyl transferase. After 1989, all patients underwent immunophenotyping. B-precursor ALL was diagnosed when CD19 was expressed by more than 20% of blasts and T-lineage ALL when 20% or more of blasts expressed CD2, CD5, and/or CD7. High-risk features of the five patients with lymphoblastic lymphoma included bone marrow involvement in four and a lactate dehydrogenase level more than 500 U/L (normal, < 185 U/L) in one.
After achieving complete remission, patients were considered high-risk if they had any of the following features: adverse cytogenetics defined as t(9,22) or t(4,11), requirement of two courses of treatment to achieve remission, or B-precursor patients with WBC counts more than 100,000/μL. These patients were taken off study for transplantation when possible.
Treatment
Treatment consisted of a total of seven courses given in the order 1A, 1B, 1C, 2A, 2B, 2C, and 3C followed by maintenance chemotherapy (Table 1). Induction chemotherapy (course 1A) consisted of daunorubicin, vincristine, prednisone, and asparaginase (DVPAsp). Patients not in remission at the end of this therapy received the first intended postremission treatment course (1B) with high-dose cytarabine (HDAC) plus etoposide. Patients not in remission at the end of the second treatment course were taken off study.
Treatment Regimen
Postremission therapy included six courses of consolidation chemotherapy. Courses 2 (1B) and 5 (2B) consisted of HDAC plus etoposide. Courses 3 (1C), 6 (2C), and 7 (3C) consisted of high-dose methotrexate plus 6-mercaptopurine. Methotrexate was given as a 220-mg/m2 intravenous bolus followed by 60 mg/m2/h (adjusted according to renal function) for 36 hours, followed immediately by leucovorin 50 mg/m2 intravenously every 6 hours for three doses, and then oral leucovorin until the serum methotrexate concentration fell below 0.05 μmol/L. Course 4 (2A) consisted of DVPAsp. At the end of the first seven courses of intensive therapy, patients received maintenance chemotherapy with oral methotrexate 20 mg/m2 weekly and oral 6-mercaptopurine 75 mg/m2 daily until they were in complete remission for 30 months. At that point, all treatment was stopped.
CNS prophylaxis included six total doses of intrathecal methotrexate (12 mg). The first dose was given at the start of induction therapy during the initial diagnostic lumbar puncture. Five subsequent doses began with the first course of postremission chemotherapy and were delivered weekly as tolerated. Patients with CNS disease at diagnosis received intensified CNS therapy. The total number of doses of intrathecal methotrexate was increased to 10 and cranial irradiation 18 Gy was given after bone marrow remission was achieved.
The first four patients on study received two additional courses of chemotherapy before the protocol was amended. This additional treatment consisted of DVPAsp (3A) and cytarabine plus etoposide (3B). Cycles 3A and 3B were eliminated because of poor hematopoietic tolerance.
Supportive Care
All patients received prophylaxis for Pneumocystis carinii with trimethoprim/sulfamethoxazole 160 mg bid twice weekly from the start of therapy until the completion of maintenance therapy. Other antibiotic prophylaxis and treatment was left to the discretion of the treating physician. Most patients received antibiotic prophylaxis against gram-negative organisms during the time when the absolute neutrophil count (ANC) was less than 500/μL and most patients received prophylaxis against fungal infections with either low-dose amphotericin or fluconazole during the period of neutropenia. Granulocyte colony-stimulating factor (G-CSF) was not used routinely during the early years of this study but was used starting in 1995, a time after which half of the patients in this study were treated. G-CSF was given in a dose of 5 μg/kg subcutaneously daily starting on day 14 of chemotherapy and continuing until the ANC was more than 1,500/μL for 2 days.
Criteria for Response
Patients were considered to be in CR when the ANC was more than 1,000/μL, platelets were more than 100,000/μL, bone marrow morphology was normal with less than 5% blasts, there was no evidence of extramedullary leukemia, and there was resolution of previously abnormal cytogenetics.
Statistical Analysis
Event-free survival (EFS) of all patients entered onto the study was calculated from initiation of therapy until an adverse event, either induction failure, relapse, or treatment-related death. Sixteen patients were taken off study and censored. These included 13 patients who received transplantation, two patients with major protocol violations, and one patient who died in an automobile accident. EFS of remission patients was calculated from the time of remission until relapse or treatment-related death. Patients were censored when taken off study as described above. Overall survival (OS) was calculated from study entry and no patients were censored. DFS of remission patients was calculated from time of remission until relapse. Data were analyzed as of March 15, 2001. EFS, DFS, and OS were calculated by Kaplan-Meier analysis and 95% confidence intervals were calculated. Comparison of outcomes between patient groups was performed using the Cox-Mantel method.
For patients with B-precursor disease, a high-risk subgroup was defined as having one or more of the following: adverse cytogenetics with t(9,22) or t(4,11), requirement for more than one treatment course to achieve complete remission, or WBC count more than 100,000/μL. Patients with T-lineage disease were considered high-risk only if they required more than one treatment course to achieve remission.
RESULTS
Patients
The median age of the 84 patients enrolled was 27 years (range, 16 to 59 years), and eight patients were older than age 50 (Table 2). The median WBC count was 9,000/μL (range, 1,000 to 660,000/μL) and 16 patients (19%) had a WBC count more than 100,000/μL. Eight patients (10%) had evidence of CNS leukemia at the time of diagnosis.
Patient Features
Immunophenotype was determined in 80 (95%) of 84 patients. Eighteen patients (21%) had T-cell disease. Their median age was 29 years (range, 18 to 49 years) and six patients were older than age 40. A mediastinal mass was present in eight (44%). CNS disease was present at diagnosis in four patients, and five patients had a WBC count more than 100,000/μL. Sixty-two patients (74%) had B-precursor disease. Their median age was 28 years, with nine patients older than age 40. Four patients had CNS disease at diagnosis. Nineteen B-lineage patients were designated high-risk at diagnosis on the basis of adverse cytogenetics in 15—nine with t(9,22) and six with t(4,11)—and WBC count more than 100,000/μL in four.
Cytogenetic evaluation was not performed in 19 patients and did not yield assessable metaphases in another seven. Of 58 patients with adequate cytogenetic analysis, 22 had a normal karyotype, nine had the Philadelphia chromosome, six had t(4,11), and 21 had other abnormalities.
Remission Induction
CR was achieved in 78 (93%) of 84 patients. There was one treatment-related death, and five patients had resistant disease. Remission was achieved with the first course of treatment (1A) in 74 patients and required the second treatment course (1B) in four.
The ANC recovered to more than 500/μL in a median of 18 days (range, 1 to 31 days) and to more than 1,000/μL in a median of 23 days (range, 14 to 35 days) (Table 3). Six patients took longer than 30 days to reach ANC more than 500/μL and 14 required more than 30 days to reach ANC more than 1,000/μL. The duration of neutropenia (ANC < 500/μL) was a median of 14 days (range, 0 to 30 days). Platelets recovered to more than 20,000/μL, more than 50,000/μL, and more than 100,000/μL at median times of 15, 18, and 22 days, respectively.
Hematologic Toxicity (days to recovery)
Nonhematologic toxicity was modest. There was little hepatotoxicity, with a median peak bilirubin of 1.4 mg/dL (range, 0.3 to 33 mg/dL). Only 11 patients had a peak bilirubin more than 3 mg/dL and only one had a peak bilirubin more than 10 mg/dL. There was minimal mucosal toxicity. Only eight patients required parenteral nutrition, and two required narcotic analgesia.
Postremission Therapy
The second (1B) and fifth (2B) courses of treatment with HDAC and etoposide produced considerable myelosuppression (Table 3). Time to recovery of ANC more than 1,000/μL took a median of 20 days, and patients spent a median of 12 days with ANC less than 500/μL (Table 2). Platelet recovery to more than 20,000/μL required 19 to 22 days. Patients required a median of three to four platelet transfusions and 6 to 7 units of RBCs during these two courses. Median duration of hospitalization was 22 and 20 days, respectively. There were no treatment-related deaths during these courses. There was little hepatotoxicity. The median peak bilirubin was 0.9 to 1.1 mg/dL. During each course, only one patient experienced a peak bilirubin more than 3 mg/dL. Although most patients did not require parenteral nutrition or narcotic analgesia, this treatment did produce significant mucositis and enteritis in some patients. During course 1B, 15 patients required parenteral nutrition and seven patients required this level of support for more than 1 week. Ten patients required narcotic analgesia and five required this for more than 1 week. During course 2B, five patients required parenteral nutrition, two for longer than 1 week, and seven patients required narcotic analgesia, six for more than 1 week. It is likely that the greater mucosal toxicity during course 1B (compared with 2B) is related to the concurrent administration of intrathecal methotrexate during this course.
Consolidation 2A included DVPAsp (Table 1). The daunorubicin dose was the same as induction and the cumulative dose of asparaginase was the same as induction, although the schedule was altered to allow outpatient administration. Vincristine and prednisone were reduced to 3 weeks. Neutrophils recovered to more than 500/μL and 1,000/μL at 16 and 17 days, respectively. The total duration of neutropenia (days with ANC < 500/μL) was 5 days, significantly shorter than during induction therapy, presumably because patients began this therapy with good marrow function. Although the intent was to deliver this treatment on an outpatient basis, the median number of days of hospitalization was 7. There was a trend toward a higher likelihood of remaining out of the hospital after G-CSF was added as a supportive measure in 1995. There was little nonhematologic toxicity during this treatment. Only one patient required parenteral nutrition and only one required narcotic analgesia. The median peak bilirubin was 1.4 mg/dL. Only four patients had a bilirubin level more than 3 mg/dL, and it rose to more than 10 mg/dL in only one. There were two treatment-related deaths during this treatment course. Both were because of sepsis and were related to delayed administration of antibiotics.
Treatment with high-dose methotrexate and 6-mercaptopurine was well tolerated. Although the therapy was administered in the hospital in all but one patient, none required hospitalization for treatment of complications. A total of 356 infusions of high-dose methotrexate were given. Only two of these courses were complicated by significant mucositis, and this never lasted longer than 2 days. Only three courses were complicated by bilirubin more than 3 mg/dL, and this resolved without intervention. The median peak serum methotrexate level at the end of infusion was 28 μmol/L. Target methotrexate levels of 20 μmol/L were reached in most patients. Only 22 patients (6%) had methotrexate levels less than 20 μmol/L, and this was less than 15 μmol/L in six (2%). Levels exceeded 40 μmol/L in only 16 infusions (4%).
Overall, patients spent a median of 67 days in hospital to receive postremission therapy. Forty-nine of these days were related to the more intensive courses and 18 days were for the purpose of administration of high-dose methotrexate chemotherapy.
Maintenance chemotherapy was generally well tolerated, and there were no serious complications. Doses of methotrexate and 6-mercaptopurine were adjusted to try to avoid significant symptoms and to keep the ANC above 1,000/μL and platelets above 75,000/μL.
Outcome
Seventy-eight of 84 patients (93%) achieved CR (Table 4). Of these 78, 13 patients with high-risk features were taken off study for transplantation, two patients had major protocol violations (described below), and one patient died in an automobile accident. Of the 62 remaining patients who received protocol therapy, there were two treatment-related deaths and 26 relapses. With median follow-up of 5.6 years (range, 1.6 to 13.3 years), 5-year EFS for patients achieving remission is 52% ± 13% (Fig 1) and 5-year DFS is 54% ± 13%. At 5 years, EFS of all patients entered onto the study is 48% ± 13% and OS of all study patients is 47% ± 13%.
Outcome of Therapy
Fig 1. EFS of all patients who entered CR. Error bars indicate 95% confidence intervals.
Of 22 high-risk patients, 13 were taken off study for transplantation. Six patients were treated with allogeneic transplantation, two with B-lineage ALL and WBC count more than 100,000/μL, one requiring two courses of induction to achieve remission, two with t(9,22), and one with t(4,11). Of these six, four remain in remission from 1.8 to 4.9 years. Seven patients were treated with autologous stem-cell transplantation, one with B-lineage ALL and WBC count more than 100,000/μL, three requiring two courses of induction to achieve remission, two with t(4,11), and one with t(9,22). Of these seven, three remain in remission from 1.2 to 3.2 years. All nine high-risk patients who did not undergo transplantation relapsed, with median time to relapse of 2.5 months (range, 1 to 9 months).
Fifty-six patients lacked high-risk features and were designated as standard risk. Fifty-three received intended protocol therapy (two major violations and one death in an automobile accident). Of these 53 patients, there were two treatment-related deaths and 17 relapses. The median time to relapse was 1.9 years (range, 0.4 to 4.9 years). One late event at 4.9 years was a case of secondary acute myelogenous leukemia. This patient is now alive and well 4.6 years after an allogeneic transplant, which represents the only successful salvage of a patient who relapsed in this study. The next latest relapse was seen at 3.9 years. With median follow-up of 5.6 years, EFS in this standard-risk patient group is 60% ± 14%.
Immunophenotype was not as strong a prognostic factor as in our previous study. For patients with T-cell disease, 5-year EFS was 48% ± 30%. However, with these small patient numbers, these results are affected by the fact that both treatment-related deaths and the one case of secondary acute myelogenous leukemia occurred in T-cell patients. For patients with B-lineage disease, outcome depended on risk status. As noted above, all high-risk patients relapsed unless treated with transplantation. For 39 patients with standard-risk B-precursor disease, 5-year EFS (and DFS) was 66% ± 16%, significantly better than in high-risk patients (P = .0008). Among the standard-risk group, 21 had especially favorable features with both age younger than 30 years and WBC count less than 30,000/μL. EFS (and DFS) in this favorable subgroup was 78% ± 21% as opposed to 52% ± 25% in the 18 remaining patients (P = .10) (Fig 2).
Fig 2. EFS of favorable (FAV), intermediate (INT), and poor-risk (POOR) B-lineage patients. Error bars indicate 95% confidence intervals.
As expected, outcome was affected by age. Patients younger than age 30 had projected EFS of 63% ± 16%, as opposed to 33% ± 22% for older patients. The outcome of the small number of patients with high-risk lymphoblastic lymphoma was not different from those with ALL. Two of these five patients remain in continuous CR, and another has been in remission (4.6 years) after allogeneic transplant in second remission. One patient failed to achieve remission and one died early in remission in an automobile accident.
There were no cases of isolated CNS relapse and only one case of concurrent bone marrow and CNS relapse. The actuarial risk of CNS relapse is 3%. However, it should be noted that the two cases of major protocol violation for which patients were censored involved failure to administer appropriate CNS prophylaxis. One patient never received CNS prophylaxis after achieving remission and developed a CNS relapse followed by systemic relapse. A second patient with CNS disease at diagnosis received only six doses of intrathecal methotrexate and did not receive either cranial irradiation or extended intrathecal therapy. This patient had a concurrent systemic and CNS relapse.
Comparing the results of the current study (protocol 8707) with those using our previous chemotherapy (protocol 8001), 5-year EFS for all remission patients is 52% v 36% (P = .11) (Fig 3). For high-risk patients, the results are the same, with no long-term disease-free survivors in patients who did not undergo transplantation. For T-cell patients, the outcomes are similar, with 5-year EFS of 48% v 55%. For standard-risk patients with B-precursor disease, the results seem significantly improved, with 5-year EFS of 66% v 34% (P = .01) (Fig 4). For all B-precursor patients (including those with high-risk disease), 5-year EFS is 55% v 27% (P = .03).
Fig 3. Comparison of EFS of all remission patients between former study (protocol 8001) and current study (protocol 8707). Error bars indicate 95% confidence intervals.
Fig 4. Comparison of EFS of standard-risk B-lineage patients between former study (protocol 8001) and current study (protocol 8707). Error bars indicate 95% confidence intervals.
DISCUSSION
This treatment program seems to have fulfilled our original expectations. The CR rate remains high (93%) as in our previous study. The ability to achieve high remission rates has been observed in other single-institution and cooperative group studies. Hyper–cyclophosphamide, vincristine, doxorubicin, and dexamethasone therapy has produced a remission rate of 94%, and anthracycline intensive therapy has produced a remission rate of 93%.10,11 The Cancer and Leukemia Group B has achieved CR rates of 89% and 87% in consecutive studies.3,12 A complete remission rate of 84% has been achieved with a novel treatment approach using HDAC and mitoxantrone and without the use of vincristine or prednisone.13 However, other groups have reported lower rates for remission induction.14,15 Whether these differences are related to treatment variables of patient selection or supportive care modalities remains to be determined.
The current treatment program was successful in avoiding some of the toxicities seen in our previous study. No cases of cardiomyopathy were observed. This would be expected given the maximal cumulative daunorubicin dose of 420 mg/m2. Anthracycline dose intensity may be an important feature in improving outcomes in adult ALL. One group has shown that treatment courses with daunorubicin doses more than 175 mg/m2 were associated with improved outcomes compared with treatments with lower anthracycline doses.11,16 Each of our two daunorubicin-containing courses of treatment use doses of at least 180 mg/m2. Because daunorubicin is the only significant myelosuppressive agent used in the DVPAsp regimen, it may be possible to escalate the dose further without prohibitive toxicity, and this may still be done keeping the total cumulative anthracycline dose at a range that would make cardiotoxicity unlikely.
The omission of cranial irradiation from our CNS prophylaxis regimen did not lead to a high risk of CNS relapse. However, the regimen was specifically designed to include repeated high systemic doses of methotrexate, achieving serum methotrexate concentrations more than 20 μmol/L, which should produce therapeutic CSF levels. The high systemic doses of cytarabine and etoposide used may also have contributed to CNS prophylaxis. Other groups have demonstrated that cranial irradiation can be safely omitted, provided that specific attention is paid to giving systemic therapy with CNS penetration. The hyper–cyclophosphamide, vincristine, doxorubicin, and dexamethasone regimen resulted in a 3% risk of CNS relapse without the use of cranial irradiation.17 Other groups have used repeated doses of intraventricular therapy through an Ommaya reservoir with satisfactory control of CNS disease.5
This new shortened and intensified treatment regimen appears to have improved outcome in some groups of patients. The increase in 5-year DFS from 35% to 52% is not statistically significant. However, this aggregate group includes high-risk patients for whom modification of nonablative therapy is not expected to improve outcome. In both of our studies, all patients with these high-risk features who did not undergo transplantation relapsed, and the time to relapse was short. Allogeneic transplantation has been shown to be potentially curative for these high-risk patients with ALL, including those with Philadelphia chromosome–positive ALL and those with primary induction failure.18,19 Adults with t(4,11) also appear to benefit from treatment with transplantation.20 Allogeneic transplantation should be considered the treatment of choice for such high-risk patients, provided they are of suitable age and have a suitable donor. For patients who seem inevitably destined to relapse with conventional approaches and who are not candidates for allogeneic transplant, autologous transplantation may offer another potentially curative approach. New aggressive approaches to autologous transplantation including intensive pretransplant cytoreduction, in vitro purging, and intensified preparative regimens have produced prolonged DFS in 25% of patients with Philadelphia chromosome–positive ALL.21 A similar treatment approach has yielded preliminary results, with 2-year DFS of 46% in patients with high-risk ALL.22 A retrospective analysis of autologous transplantation reported to the Autologous Bone Marrow Transplant Registry compared to unrelated donor, and allogeneic transplantation reported to the National Marrow Donor Program, has suggested that autologous transplantation is a reasonable treatment option for patients with high-risk ALL.23
Although we thought it unlikely that this modified and intensified therapy would benefit the highest risk group of patients, we hoped that this new treatment approach would improve outcomes for patients with standard-risk B-precursor disease. There was a need for significant improvement because, in our previous study, 5-year DFS was only 34% in this patient group. The major change in therapy was the increased reliance on antimetabolites. High-dose methotrexate was given early and repeatedly for a total of six courses. In addition to providing CNS prophylaxis, high-dose methotrexate may have contributed to the improved outcome in B-precursor patients. High-dose methotrexate has been shown to be an important feature of successful treatment regimens for mature B-cell ALL with the suggestions that high doses of 8 g/m2 per dose may be superior to 3 g/m2.24,25 The success of many pediatric programs in treating B-precursor ALL may be related to the heavy reliance on antimetabolite therapy. However, the role of high-dose methotrexate in changing outcomes is speculative, as there were many other changes in this treatment program.
In conclusion, we report that this program of intensified cyclical chemotherapy produces a high rate of CR for adults with ALL, and that patients without high-risk features have a 60% chance of remaining disease-free at 5 years. Adequate protection from CNS relapse can be provided without cranial irradiation by the use of high-dose systemic chemotherapy combined with intrathecal methotrexate. However, for patients at high risk of relapse, it is unlikely that modification of nonablative chemotherapy will alter their prognosis, and these patients should be treated with either allogeneic or autologous transplantation whenever possible.
- Received July 24, 2001.
- Accepted February 11, 2002.
