See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/15480063

A Five-Drug Remission Induction Regimen With Intensive Consolidation for

Adults With Acute Lymphoblastic Leukemia: Cancer and Leukemia Group B
Study 8811

Article  in  Blood · April 1995

DOI: 10.1182/blood.V85.8.2025.bloodjournal8582025 · Source: PubMed

CITATIONS READS

496 1,139

10 authors, including:

Richard Larson Stanley R Frankel

University of Chicago Columbia University
852 PUBLICATIONS   66,188 CITATIONS    136 PUBLICATIONS   13,406 CITATIONS   

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Stanley R Frankel on 21 May 2014.

The user has requested enhancement of the downloaded file.

A Five-Drug Remission Induction Regimen With Intensive Consolidation for
Adults With Acute Lymphoblastic Leukemia: Cancer and Leukemia
Group B Study 8811
By Richard A. Larson, Richard K. Dodge, C. Patrick Burns, Edward J. Lee, Richard M. Stone, Philip Schulman,
David Duggan, Frederick R. Davey, Robert E. Sobol, Stanley R. Frankel, Arthur L. Hooberman, Carol A. Westbrook,
Diane C. Arthur, Stephen L. George, Clara D. Bloomfield, and Charles A. Schiffer

The goal of this phase II multicenter clinical trial was t o tion, and maintenance chemotherapy for a totalof 24
evaluate a new intensive chemotherapy program for adults months. After a median follow-up time of 43 months, the
with untreated acute lymphoblastic leukemia (ALL) and t o median survival for all197 patients is 36 months; the median
examine prospectively the impact of clinical and biologic remission duration for the 167 CR patients is 29 months.
characteristics on the outcome. One hundred ninety-seven Favorable pretreatment characteristics relative t o remission
eligible and evaluable patients (16 t o 80 years of age; me- duration or survival are younger age, the presence of a medi-
dian, 32 years of age) received cyclophosphamide, daunoru- astinal mass or lymphadenopathy, a white bloodcell count
bicin, vincristine, prednisone, and L-asparaginase; 167 pa- (WBC) lessthan 3O,OOO/pL, L1 morphology, T or TMy immu-
tients (85%) achieved a complete remission (CR),13(7%) nophenotype, and the absence of the Ph chromosome. The
had refractory disease, and 17 (9%)died duringinduction. A estimates of the proportionsurviving at 3 yearsare 69% for
higher CR rate was observed in younger patients (94% for patients less than 30 years old, 39% for those 30 t o 59 years
those c 3 0 years old, 85% for those 30 t o 59 years old, and old, 89% for those who had a mediastinal mass, 59% with
39% for those r60 years old, P < .001) and in those who WBC less than 30,OOO/pL, 63% with L1 morphology, 69% for
had a mediastinal mass (100%) or blastswith a T-cell immu- T or TMy antigen expression, and 62%for those who lack the
nophenotype. Eighty percent of B-lineage and 97% of T-cell Ph chromosome. Fifteen patients (8%) had no unfavorable
ALL patients achieved a CR ( P = .01). The coexpression of prognostic factors and have an estimatedprobability of sur-
myeloid antigens did not affect the response rate or dura- vival at 5 years of 100% (95% confidence interval, 77% to
tion. Seventy percent of those with cytogenetic or molecular 100%). This intensive chemotherapy regimen produces a
evidence of the Philadelphia (Ph) chromosome and 84% of high remission rate and a high proportion ofdurable remis-
those without such evidence achieved a CR ( P = .11). Pa- sions in adults with ALL.
tients in remission received multiagent consolidation treat- 0 7995 by The American Society of Hematology.
ment, central nervous system prophylaxis, late intensifica-

A LTHOUGH RECENT clinical trials have shown that

65% to 85% of adults with acute lymphoblastic leuke-
mia (ALL) may achieve a complete remission (CR), these
mens extending over a number of months or years may offer
the greatest likelihood for producing long-term disease-free
survival for adults with ALL.”
remissions have been disappointingly short, especially for The goal of this study was to evaluate a new intensive
older adults.”’ Recent treatment strategies have focused on chemotherapy program for adults and to examine prospec-
increasingly intensive induction and postremission treatment tively the impact of clinical and biologic characteristics on
with multiple chemotherapy agents and also on multivariate the outcome. This Cancer and Leukemia Group B (CALGB)
analyses of prognostic factors to individualize therapy. study of adults with ALL tested the feasibility of the direct
Among the clinical and biologic characteristics that have
been found to have prognostic importance for adults with
ALL are age, initial white blood cell count ( W C ) , platelet From the University of Chicago, Chicago, IL; the CALGB Statisti-
count, serum lactate dehydrogenase and albumin levels, the cal Center, Durham, NC; the University of Iowa, Iowa City, IA; the
presence of adenopathy or hepatosplenomegaly, perfor- University of Maryland Cancer Center, Baltimore, MD; the Dana
mance status, immunophenotype, cytogenetics, and time to Farber Cancer Institute, Boston, MA;the North Shore University
achieve CR.”5 Hospital, Manhasset, NY; the State University of New York Health
Science Center at Syracuse, NY; the University of California at San
Investigators at single institutions have reported favorable
Diego, CA; the Roswell Park Cancer Institute, Buffalo,NY;the
results using intensive and prolonged remission consolida- University of Minnesota, Minneapolis, MN; and CALGB, Lebanon,
tion programs. At the Memorial Sloan-Kettering Cancer NH.
Center, multiple courses of eight drugs were used in various Submitted September 27, 1994; accepted November 28, 1994.
combinations in the L-l0 and L-1OM protocols; a CR rate Supported in pari by National Institutes of Health Grants No.
of 85% was reported among 72 patient^.^.' The median re- CA31946, CA33601, CA37027, CA37055, CA41287, and AGO353
mission duration was 51 months, and disease-free survival and by the Coleman Leukemia Research Fund.
at 5 years was estimated to be 45%. However, when evalu- Presented in part at the 25th Annual Meeting of the American
ated in a multi-institutional trial, the same chemotherapy Society of Clinical Oncology, San Diego, CA, May, 1992.
program was reported to yield a CR rate of 68% in 168 Address reprint requests to Richard A. Lurson, MD, university
patients; the median duration of CR was 23 months.’ A of Chicago Medical Center, 5841 S Maryland Ave, MC2115, Chi-
cago, IL 60637.
similar program tested in 59 consecutive patients at the Uni-
The publication costs ofthisarticle were defrayed in part by page
versity of Iowa showed a CR rate of 75%; the median dura- charge payment. This article must therefore be hereby marked
tion of remission was about 50 months.’ Fifteen patients “advertisement” in accordance with 18 U.S.C. section 1734 solely to
(34%) remained in continuous remission longer than 5 years. indicate this fact.
These results and others have suggested that additional initial 0 1995 by The American Society of Hematology.
cytoreduction followed by multiagent intensification regi- oooS-4971/95/8508-0003$3.00/0

Blood, Vol 85, No 8 (April 15), 1995: pp 2025-2037 2025

2026 !ARSON ET AL

application of treatment programs that have proven to be among the other B-lineage cases in subsequent analyses. Patients
very effective in high-risk childhood ALL. To achieve more with myeloperoxidase-negative blasts that expressed only myeloid
rapid cytoreduction, our protocol used a five-drug induction antigens (and not B- or T-lymphoid antigens) were reclassified as
regimen that was derived by adding a single dose of cyclo- acute myeloid leukemia (AML), subtype MO, and were deemed to
be ineligible (4patients).
phosphamide (1,200 mg/m2) to a slight modification of the
Treatment protocol. The drugs and dosages used in the induc-
four-drug regimen used in two previous CALGB studies tion, consolidation, and maintenance phases of treatment are listed
(7612 and 801 A similar five-drug induction program, in Table 1. The induction course used a single dose of cyclophospha-
when used in high-risk childhood ALL (Children's Cancer mide on day 1, 3 consecutive days of daunorubicin, weekly vincris-
Group study CCG-l92P), yielded a CR rate of 96% with tine, biweekly L-asparaginase, and 3 weeks of prednisone. Before
only 2% deaths during ind~cti0n.I~ We modified the standard each L-asparaginase injection, the serum amylase activity was mea-
consolidation program used in the German (Berlin-Frank- sured. In addition, it was recommended that fresh frozen plasma or
furt-Munster [ B M ] ) multicenter trial for adults by increas- cryoprecipitate be transfused to keep the fibrinogen level greater
ing the dose of cyclophosphamide from 600 to 1,000 mg/ than 100 mg/dL. Initially, there were no dose reductions for older
m2 and adding 2 weeks of vincristine and L-asparaginase patients. After 1 year of accrual to the study (76 patients), one-third
dose reductions were implemented for patients older than 60 years
treatment during the expected period of myelosuppression.'
for the cyclophosphamide and daunorubicin and the prednisone ther-
To maintain maximum dose intensity, no dose reductions apy was shortened to 7 days, because of a high induction death rate
nor delays were permitted for myelosuppression in the ab- caused by infection in this age group.
sence of fever or infection. In addition, earlier and more Early intensification (course 11) included 2 months of treatment
extensive use of L-asparaginase was prescribed in this proto- using cyclophosphamide, subcutaneous cytarabine, oral 6-mercapto-
col than has commonly been used in adult ALL treatment purine (6-MP), vincristine, and more subcutaneous L-asparaginase.
programs (biweekly administration for 7 weeks during the Two years after the study opened (156 patients), the protocol was
first 12 weeks of therapy). amended so that CNS prophylaxis was initiated with the first two
doses of intrathecal methotrexate during course 11, rather than as
previously performed in course III, to provide earlier CNS prophy-
MATERIALS AND METHODS laxis.
Patients. Patients were eligible if they had untreated ALL of In course 111, the CNS prophylaxis was completed with cranial
any of the three French-American-British (FAB) subtypes or acute irradiation (2,400 cGy) and 5 weekly doses of intrathecal methotrex-
undifferentiated leukemia.I4 Patients were registered by telephone ate with daily 6-MP, followed by a brief maintenance period using
with the CALGB Statistical Office before treatment. The diagnosis daily oral 6-MP andweekly oral methotrexate. Course IV was a
of ALL was confirmed by central review of blood smears and bone late intensification course lasting 8 weeks, followed by prolonged
marrow (BM) specimens for cytologic and cytochemical features maintenance treatment with daily 6-MP andweekly methotrexate
according to the FAB riter ria.'^.'^ Central immunophenotyping and plus monthly pulses of vincristine and prednisone.
pathology review were required. It was recommended that pretreat- The total duration of treatment was 24 months. Testicular biopsies
ment blood and BM specimens be submitted for cytogenetic analysis, were not required at the end of therapy, and testicular irradiation
including central review of the karyotypes (CALGB study 8461). was not administered prophylactically. Patients who had an isolated
and for molecular analysis for the presence of the BCR-ABL fusion CNS relapse while continuing in a marrow remission were counted
gene (CALGB study 8762).",Is Leukemia cells were analyzed in a as failures; however, they continued to receive systemic chemother-
central laboratory by Southern blot and pulsed-field gel electrophore- apy on protocol after suppression of cerebrospinal fluid (CSF)
sis for rearrangements within the BCR gene according to previously lymphoblasts with additional intrathecal chemotherapy.
reported methods, using conditions that will detect both the p190 No hematopoietic growth factors were used. Cotrimoxazole or
and p210 subtypes." A lumbar puncture for spinal fluid examination aerosolized pentamidine were recommended for Pneumocystis pro-
was not recommended at diagnosis for asymptomatic patients. How- phylaxis, starting in course 111. The use of oral nonabsorbable antibi-
ever, patients with symptomatic central nervous system (CNS) leuke- otics, the management of febrile episodes and transfusions, and the
miawerenot excluded but received additional CNS therapy. All use of hospitalization were not prescribed by the protocol, but were
patients were older than 15 years, had adequate renal and hepatic rather left to institutional guidelines.
function (less than twofold elevated above the normal range unless Data audit. CALGB central data management personnel were
felt to be caused by leukemia infiltration), and hadprovided informed responsible for quality assurance for all clinical data submitted for
consent. this study. Eligibility criteria were verified for all patients and an
Immunophenotyping (CALGB study 8364) was performed in two evaluation of treatment, response, and toxicity was made by the
central CALGB laborat~ries.'~ In 10 cases in which the pretreatment study chair (R.A.L.). In addition, as part of the group data monitoring
specimen was notevaluable in the central laboratory, immunopheno- program, members of the CALGB Data Audit Committee made
typing data from the local institutions were used after central review. periodic site visits to all institutions to verifycompliance with federal
Flow cytometric analysis and a panel of monoclonal antibodies were regulations and protocol requirements, including eligibility, treat-
used for indirect immunofluorescence. The criterion for surface ment, response data, and follow-up.2oA random subset of 68 patients
marker positivity was expression by at least 20% of the leukemia (32%) treated on this study had such an on-site review of their
blast cell population. B-lineage antigen expression was defined as medical records. In addition, all radiotherapy portals and dosimetry
CD19 or CD20 positivity; T-lineage antigen expression as CD5 or records for the cranial irradiation were centrally reviewed for quality
CD2 reactivity; and myeloid antigen expression as CD13 or CD33 control by the Quality Assurance Review Center (Providence, RI).
positivity. Expression of the common ALL antigen (CALLA) was Criteriafor response. Patients were considered tobeinCR
assessed by CD10 reactivity. Cases expressing combinations of both when the neutrophil count was greater than 1,50O/pL, the platelet
B-lineage and T-lineage antigens were classified as BT, BTMy, or count was greater than 1OO,OOO/pL, the results of BM examination
miscellaneous. Cases expressing surface membrane Ig (SmIg) were were normal (including 4 % blasts and >25% cellularity), and all
considered FAB-L3 (Burkitt-type ALL) andwerenot included extramedullary disease had resolved. Patients with greater than 25%
NT INTENSIVE ALL 2027

Table 1. Chemotherapy Regimen for ALL in Adults

Course I: Induction (4wk)

Cyclophosphamide* IV Day 1,200mglm’ 1
Daunorubicin* IV 45 mg/rn2 Days 1, 2,3
Vincristine IV Days 2 mg 22 15, 1, 8,
Prednisone* Days mglm’ld
ponv 60 1-21
L-asparaginase sc 6,000 IUlm’ Days 5,22
8. 11, 15, 18,
+For patients ~ 6 yr0 old:
Cyclophosphamide 800 mg/m2 Day 1
Daunorubicin 30 mg/m2 Days 1,2,3
Prednisone Days mg/m’/d 60 1-7
Course II: Early intensification (4wk, repeat once)
Intrathecal methotrexate 15 mg Day 1
Cyclophosphamide IV 1,000rnglm’ Day 1
6-Mercaptopurine PO mglm’ld 60 Days 1-14
Cytarabine sc 75 mg/m’/d Days 1-4. 18-1
Vincristine IV 2 mg Days 15,22
L-asparaginase sc 6,000 IU/m’ Days 15, 18.22, 25
Course 111: CNS prophylaxis and interim maintenance (12wk)
Cranial irradiation 2,400 cGy Days 1-12
Intrathecal methotrexate 15 mg Days 1, 8. 15, 22, 29
6-Mercaptopurine PO 60 mghn’/d Days 1-70
Methotrexate PO 20 mg/m2 Days 36, 43, 50, 57, 64
Course IV: Late intensification (8wk)
Doxorubicin IV 30 mg/m’ Days 1, 8, 15
Vincristine IV 2 mg Days 1.8. 15
Dexamethasone PO 10 mg/m2/d Days 1-14
Cyclophosphamide IV 1,000mg/m’ Day 29
6-Thioguanine PO 60 mglm‘/d Days 29-42
Cytarabine sc 75 mg/m2/d Days 29-32, 36-39
Course V: Prolonged maintenance (until 24 rno from diagnosis)
Vincristine IV 2 rng Day 1 of every 4 wk
Prednisone PO mg/m’/d 60 Days 1-5 of every 4 wk
Methotrexate PO 20 mglm‘ Days 1,s. 15, 22
6-Mercaptopurine PO mglm’ld 60 Days 1-28

lymphoblasts remaining in the BM after course I were removed from CR rate and duration and survival. For the joint analysis of these
this protocol. All patients were required to have achieved a CR by variables, regression analyses were used. The analysis of CR rate
half-way through course I1 to remain on the study. was performed using the logistic regression model, whereas the anal-
Staristical methods. The proportion of patients achieving a CR yses ofCR duration and survival were performed using the Cox
to the induction regimen was the primary outcome measure on this proportional hazards regression These were performed
study. The duration of CR and length of survival were additional using the SAS procedures LOGISTIC and PHREG, respectively.”
outcome measures. Differences in proportions of CRs among patient For the regression analyses, all variables that attained a univariate
subgroups were analyzed using Fisher’s exact test. The duration of P value of .20 or less were considered in the variable selection
CR was defined to be the time from achieving a CR to relapse (bone process. All reported P values are nominal two-sided values unless
marrow, blood, CNS, or testicular), death, or date of last follow-up. otherwise stated. The analysis was based on all data available as of
Patients still at risk, lost to follow-up, or withdrawn for BM trans- August 4, 1994.
plant (BMT) were censored for the analysis of remission duration.
Survival was defined as the time from study entry to death or date RESULTS
of last follow-up. Probabilities of surviving and remaining in CR
were estimated by the Kaplan-Meier method.” Ninety-five percent Patient accrual. BetweenSeptember1988and June
confidence intervals (CI) for these probabilities and the median sur- 1991, 214 patients were entered onCALGBstudy 8811
vival times were obtained using the method of Simon and Lee?’ from 25 main member institutions and 41 of their affiliated
Median follow-up time was estimated by reversing the codes for the hospitals. One patient withdrew before beginning treatment
censoring indicator in a Kaplan-Meier analysis. In instances in which and 16 patients were judged ineligible after central review
the median was not defined, the estimate was reported to be greater of all data (14 with AML and
with
2 lymphocytic
than the smallest possible time. Differences in survival or remission
duration between patient subgroups were tested using the logrank
lymphoma). Thus, 197 patients were eligible and evaluable
statistic, adjusted for multiple comparisons where appr~priate.’~.” for this report.
In accordance with the study objectives, the prognostic signifi- Patient characteristics. The 197 eligible patients ranged
cance of age, WBC count, platelet count, mediastinal mass, organ- in age from 16 to 80 years, with a median age of 32 years.
omegaly, lymphadenopathy, FAB classification, immunophenotype, Eighteen patients (9%) were 60 years or older. There were
cytogenetics, and molecular analysis were assessed with respect to 124 men (63%) and 73 women. Eighteen patients were Afri-
2028 LARSON ET AL

Table 2. Results of Cytogenetics and Molecular Assays niques. When the results of both cytogenetic and molecular
for Ph Positivity in Adult ALL tests were combined, 30 patients were positive for either the
BCR Rearrangement In) Ph chromosome or BCR-ABL rearrangement (Ph-positive
ph Chromosome
Negative
Positive
Unknown
Total h )
ALL). Forty-nine patients were analyzed by both methods:
15 were positive and 29 were negative by both tests (90%
Positive 15 2 8 25 concordance). Eighty-three patients were negative by at least
Negative 3 29 28 60 one test (and not positive by the other). Thus, approximately
Unknown 2 26 04 112'
27% of the 113 patients studied had Ph-positive ALL. Forty
Total 20 57 120t 197
percent of patients with B-lineage ALL were Ph-positive,
* Includes not done (n = 81) and not evaluable (n = 31). compared with 5% of those with T-lineage ALL.
t Includes not done (n = 114) and not evaluable (n = 6).
Remission induction. One hundred sixty-seven (85%) of
197 eligible patients achieved a CR (Table 3). One hundred
twenty-three (74%) of the responders were in CR within 30
can-American, 4 were Hispanic, and 1 wasAsian. Initial days from the first treatment; 44 (26%) required more than
WBC counts ranged from 500 to 475,000/pL (median, 30 days, either because of slow recovery of marrow cellu-
17,00O/pL), platelet counts ranged from 8,000 to 557,000/ larity and blood counts in 29 patients or because additional
pL (median, 55,500/pL), and hemoglobin levels ranged from chemotherapy (ie, course 11) was required in15 patients.
4.0 to 16.6 g/dL (median, 9.6 g/&). Eighty-eight percent of remissions were achieved within 42
Thirty-four percent of patients had WBCcounts =-30,000/ days. Eight patients (4%) had no response to induction che-
pL, 13% had WBC counts greater than lOo,OOO/pL, and 6% motherapy (ie, >25% lymphoblasts persisting in the mar-
had more extreme hyperleukocytosis with WBC counts in row) and were taken off study after 4 weeks. Five patients
excess of200,OOO/pL. Fifty-six percent of patients had a (3%) had at best only a partial response and were taken off
fever or infection before chemotherapy. Only 1 patient had study after 8 weeks. Of these 13 patients with refractory
symptomatic CNS disease at diagnosis. Thirty patients disease, 5 are known to have had a Ph chromosome or BCR
(15%) had a mediastinal mass, 60 (31%) had palpable sple- rearrangement, 1 had a t(4; 1l), and 1 had Burkitt-type (L3)
nomegaly, and 47 (24%) had hepatomegaly. Palpable lymph- ALL. Despite additional treatment, all but 1 (who received
adenopathy was present in 78 patients (40%). an allogeneic marrow transplant) have died.
One hundred sixty-six patients (84%) could be classified Seventeen patients (9%) died before hematopoietic recov-
by the pathology review committee using the FAB criteria; ery from the induction chemotherapy; 1 of these was less
71 cases (43%) were L1, 87 (52%) were L2, and 8 ( 5 % ) than 30 years old (1% of the age cohort), 7 were 30 to
were L3. Morphologic subtyping could not be accomplished 59 years old (8%), and 9 were 260 years old (50%). The
in 24 cases, usually because of inadequate marrow aspirate predominant cause for induction mortality was infection with
samples, and these were considered unclassifiable acute leu- either enteric gram-negative bacteria, Streptococcus pneu-
kemia. Seven cases were not evaluable for central review. moniae or Candida species, occumng during week 2 or 3 of
Central immunophenotyping was successfully performed treatment. There were two episodes of clinically severe tu-
in 130 cases. Where these determinants were missing, local mor lysis syndrome, and l of these patients died.
institutional imrnunophenotyping data were used in 10 cases Treatment response was a function of age (Table 4).
to assess the lymphocyte surface marker profile. Excluding Eighty-two of the 87 patients (94%) less than 30 years old
the 8 patients with L3 ALL, there was surface expression of achieved a CR, compared with 78 of 92 (85%) between 30
at least one B-lineage antigen in 86 cases (61%). Thirty-nine and 59 years of age and only 7 of 18 (39%) of those 60
cases (28%) expressed at least one T-lineage antigen and 27 years and older (P< .001). Four percent to 10% of patients
cases (19%) expressed at least one myeloid (My) antigen. within each age group had either a partial remission or no
Five subsets were defined: pure B lineage, 67 cases (48%); response to induction chemotherapy. After the protocol was
BMy, 19 cases (14%); pure T lineage, 3 1 cases (22%); TMy, amended to institute a one-third reduction of the chemother-
8 cases (6%); and a miscellaneous group with other marker apy doses for patients 2 6 0 years old (see Materials and
profiles (SmIg positive, BT, BTMy, and unclassified), 15 Methods and Table l), the CR rate for this older group
cases ( I 1%). The miscellaneous group was not included in
further analyses by immunophenotype. Sixty-five percent of
the T or TMy patients had a mediastinal mass, compared
with only 4% of the non-T-lineage patients. Eighty-three Table 3. Results of Therapy
percent of patients with a mediastinal mass had a T or TMy Patients entered 214
immunophenotype. Patients eligible 197
Eighty-five (73%) of the 116 cases submitted for cytoge- Induction deaths 17 (9%)
netic analyses were evaluable after central reviewof the Refractory disease 13 (7%)
karyotypes. The Philadelphia (Ph) chromosome was identi- CR 167 (85%)
fied in 25 patients (29%). Molecular analyses for a BCR Died in remission (6%) 10
Censored for BMT in first CR (3%)
5
gene rearrangement from pretreatment blood and/or marrow
Relapsed 77 (46%)
specimens were performed in 77 cases; 20 (26%) were posi-
CCR (45%) 75
tive. Table 2 shows the results from these two analytic tech-
ENT INTENSIVE FOR ADULT ALL 2029

Table 4. Clinical and Biologic Characteristicsin Relation to CR Rates, Remission Duration, and Survival
Survival Duration Remission

CR Probability of CCR Survival of Probability

Median Median
Variable n (%) n (%l P (mo) at 3 yr (95% Cl) P (mol at 3 yr (95% Cl) P
Total 197 (0.42-0.58)
167 (85) 0.50 (0.37-0.55)
29 0.46 36
Age (yr)
<30 87 (44) c.001 36 0.51 (0.38-0.63) .2 1 >42 0.69 (0.57-0.78) <.001
30-59 92 (47) 25 0.43 (0.30-0.57) 25 0.39 (0.28-0.51)
260 18 (9) 12 0.43 (0.16-0.75) 1 0.17 (0.06-0.39)
Gender
Male 124 (63) .69 34 0.48(0.37-0.59) .56 33 0.49 (0.40-0.59) .76
Female 73 (37) 28 0.44 (0.30-0.60) 38 0.52 (0.39-0.651
Performance status
0-1 166 (84) .58 34 0.48 (0.38-0.581 .23 38 0.51 (0.43-0.60) .40
2-4 31 (16) 15 0.36 (0.18-0.57) 21 0.45 (0.26-0.66)
Fever or infection
+ 110 (56) .43 36 0.51 (0.39-0.63) .46 39 0.53 (0.43-0.64) .50
- 85 (44) 27 0.42 (0.29-0.56) 28 0.46 (0.35-0.60)
Mediastinal mass
+ 30 (15) .01 >33 0.71 (0.49-0.87) ,002 >42 0.89 (0.70-0.96) <.001
- 166 (85) 26 0.41 (0.32-0.52) 24 0.43 (0.35-0.52)
Lymphadenopathy
+ 78 140) .07 36 0.51 (0.38-0.64) .49 >38 0.60 (0.47-0.71) .04
- 117 (60) 28 0.43 (0.31-0.56) 26 0.44 (0.34-0.55)
Splenomegaly
i W (31) .20 27 0.47 (0.33-0.62) .93 32 0.50 10.36-0.63) .69
- 135 (69) 29 0.46 (0.35-0.58) 37 0.51 (0.41-0.60)
Hepatomegaly
+ 47 (24) .65 24 0.41 (0.26-0.58) 27 32 0.51(0.35-0.66) .76
147 (76) 34 0.49 (0.38-0.60) 37 0.50 (0.41-0.60)
Leukocytes
<30,000 130 (66) .m 37 0.51 (0.40-0.62) .05 244 0.59 (0.49-0.68) .M)1
230,000 66 (34) 19 0.36 (0.22-0.53) 19 0.34 (0.23-0.47)
Platelets
<50,000 81 (41) .l6 27 0.43 (0.30-0.58) .36 23 0.44 (0.33-0.57) .11
r
5 0.
0 00 115 159) 34 0.48 (0.37-0.60) 41 0.54 10.44-0.64)
FAB
L1 71 (37) .35 38 0.54 (0.39-0.68) .l6 >44 0.63 (0.50-0.74) .03
L2 87 (46) 26 0.46 (0.33-0.59) 25 0.45 (0.34-0.57)

L3 8 14) 3 0.17 (0.03-0.56) 6 0.38 (0.11-0.74)

Unclassified 24 (13) 23 0.35 (0.16-0.60) 11 0.31(0.15-0.541
lmmunophenotype
8 67 (48) .02 25 0.42 (0.27-0.581 .l4 19 0.36 10.25-0.49) ,004
T 31 (22) >za 0.57 10.37-0.76) >40 0.67 10.47-0.82)

BMY 19 (14) 27 0.38 (0.18-0.64) 27 0.47 (0.26-0.69)

TMY 8 (6) >40 0.86 (0.49-0.97) >40 0.75 (0.41-0.93)
Other 15 (11) 31 0.53 (0.28-0.76) 43 0.60 (0.36-0.80)

B or BMy 86 (61) .01 25 0.41 (0.28-0.55) .04 21 0.38 (0.28-0.50) .001

T or TMy 39 (28) >32 0.63 (0.44-0.78) >40 0.69 (0.51-0.82)

BMy or TMy 27 (19) >31 0.55 (0.33-0.75) 233 0.55 (0.36-0.73)

Cytogenetics
Ph' 25 (29) .l5 10 0.14 (0.05-0.34) ,006 11 0.19 (0.08-0.37) ,008
Ph- W (71) >33 0.53 (0.39-0.68) 34 0.50 (0.37-0.63)
Molecular
BCR-ABL' 20 (26) .l1 7 0.10 (0.03-0.30) c.001 11 0.13 (0.05-0.33) <.m1
BCR-ABL- 57 (74) 234 0.64 (0.48-0.781 >39 0.63 (0.49-0.75)
Cytogenetics and molecular
Ph+ or BCR-ABL' 30 (27) .l 1 7 0.11 (0.04-0.28) 1.001 11 0.16 (0.07-0.32) <.001
Negative by 1 test+ 83 (73) 233 0.56(0.434.69) 44 0.56 (0.45-0.67)
Ph' or BCR-ABL+ 30 (51) .2 1 7 0.1 1 (0.04-0.28) <.W1 11 0.16 (0.07-0.32) <.W1
Negative by both tests 29 (49) >40 0.72 (0.51-0.86) >40 0.62 (0.44-0.78)
Time to CR (d)
530 123 (74) 37 0.51 (0.40-0.62) 24 244 0.59 (0.49-0.69) .40
>30 44 (26) 26 0.33 (0.20-0.50) 39 0.56 (0.40-0.71)
Cases evaluated by cytogenetics or molecularmethods but not by both tests.

changed from 3 of 10 to 4 of 8 because of fewer induction or the presence or absence of fever or infection, splenomeg-
deaths (6 [60%1v 3 [38%1, respectively). aly, or hepatomegaly (Table 4). The presence of a rnediasti-
Subset analyses. The remission induction rate did not nal mass or lymphadenopathy were favorable features asso-
vary significantly by gender, performance status (0-1 v 2-4), ciated with CR rates of 100% and 91%, respectively. The
2030 LARSON ET AL

30 years at diagnosis are estimated to remain in CCR at 3

years, compared with 43% of middle-aged and 43% of older
patients, but these differences are not statistically significant.
Remission durations were longer for patients presenting
with a WBC less than 30,00O/pL( P = .05). Remission dura-
tion was not associated with the initial platelet count but
was significantly longer in those with a mediastinal mass ( P
= .002).
Biologic features were correlated with outcome. The esti-
0.0 I
mated probability of CCR at 3 years is 54% for patients with
o 1 2 3 4 5 6 L1 morphology, 46% for L2, and 17% for L3 ( P = .l6 for
rears L1 v L2). The median remission durations are 38, 26, and
Fig 1. The duration of complete remission is shown for the 167
3 months, respectively. Sixty-three percent of those with
CR patients after a median follow-up
time of 3.5 years. The estimated expression of T-cell antigens remain in CCRat 3 years,
probability of remaining in continuousCR at 3 years is 46% (95% Cl, compared with 41% of those expressing B-lineage antigens
37% to 55%). ( P = .04). Patients withmyeloidantigen expression have
had an intermediate outcome overall with 55% in CCR at 3
years, but they do not differ from the patients without my-
presenting platelet count (greater than or less than 50,000/ eloid antigen expression within the same lymphoid lineage.
pL) did not affect the CR rate. Patients with a WBC less Only 11% of patients with Ph-positive (or BCR+) ALL
than 30,OOO/pL had a higher CR rate (88%) than those with were estimated to remain in CCR at 3 years (95% CI, 4%
a greater WBC count (77%), but this difference was not to 28%), compared with 56% (95% CI, 43%to 69%) of
statistically significant ( P = .06). The CR rate was 80% for those who were negative byat least one genetic test and
the 25 patients with initial WBC counts greater than 100,000/ 72% (95% CI, 51% to 86%) of those who were negative by
pL and 73% for the 11withWBC counts greater than both cytogenetic and molecular analyses ( P < .001 for each
200,OOO/pL. The CR rate did not vary according to the pre- comparison).
treatment marrow cellularity or the fraction of lymphoblasts Fifty-one percent of patients who achieved a CR within
in the marrow. The CR rate was 90% for patients with FAB 30 days were estimated to remain in CR after 3 years, com-
L1 morphology, 84% for L2, and 75% for L3 ( P = .35 for pared with 33% of those who needed longer than 30 days
L1 v L2). to enter remission, but this difference was not statistically
There were significant differences in the CR rates ac- significant ( P = .24). Similarly, those 20 patients who re-
cording to immunophenotype and genetic subtype. One hun- quired more than 42 days to achieve CR have not had sig-
dred percent of 3 1 patients with a pure-T-lineage phenotype nificantly shorter remission durations (median, 26 months v
achieved a CR, compared with 82%of the 67 with a pure-B- 29 months for rapid responders, P = .62).
lineage phenotype ( P = .02). The coexpression of myeloid After adjusting for age in a logistic regression model, the
antigens had no impact on the CR rate, ie, 78% for these 27 WBC count retained statistical significance with respect to
patients overall (88% for 8 TMy patients and 74% for 19 CR duration ( P = .006). After controlling for these two
BMy patients). There were no differences in these results factors, mediastinal mass was the most significant variable
when only the cases with central immunophenotyping were with respect to remission duration ( P < .001). After ad-
analyzed. justing for all three factors, FAB subtype (L1 v L2) attained
The CR rate was 68% for the 25 patients with a t(9;22) prognostic significance for CR duration ( P = .005), as did
by cytogenetic analysis and 65% for the 20 patients in whom immunophenotype (T or TMy v others; P = .04), whereas
a BCR-ABL rearrangement was detected. One hundred thir- adverse cytogenetics [the detection of Ph, or the BCR/ABL
teen patients were studied by at least one genetic method. gene, or a t(4; 1l)] retained significance ( P = M ) . Among
The CR rate was 70% for the 30 patients identified by one T or TMy patients, the WBC count did not have prognostic
or both tests as having Ph-positive ALL and 84% for those significance for CR duration, but a mediastinal mass was
who were negative ( P = .11). significantly associated with longer remission duration ( P =
Using a logistic regression model and adjusting for age, .04).
the lack of T-cell antigen expression ( P = .03) and the Overallsurvival. The estimated median survival for all
absence of a mediastinal mass ( P = .04) were indicators of 197 treated patients is 36 months after a median follow-up
a poor remission induction outcome. No other factors were time of 43 months (range, 24 to 64 months; Fig2). The
related to the response rate after adjusting for age. median survival for the 167 patients who achieved a CR has
Remission duration. The remission duration estimates not been reached but is longer than 45 months. The survival
for the 167 patients who achieved a CR are shown in Fig 1. on this study is significantly better than that observed on
The median duration of CR for all responders is 29 months any of the three previous CALGB studies ( P < ,001 for all
after a median follow-up time of 40 months (range, 23 to 3 comparisons); CALGB 8513 had a median survival of 19
63 months). Approximately 46% of all responders are esti- months, and CALGB studies 7612 and 8011 had medians
mated to remain in continuous CR (CCR) at 3 years (95% of 16 months (Fig 3).".'*,**
CI, 37% to 55%). Fifty-one percent of patients younger than Patient age is significantly associated with survival ( P <
INTENSIVETREATMENT FOR ADULT ALL 2031

N
0.2 0.2- I 60. years(n-18)

0.0 0.oi (.

0 1 2 3 4 5 6
Years years

Fig2. The survivalof all 197eligiblepatientsisshown after a Fig 4. The survival of the 87 patients less than 30 years old (me-
median follow-up time of 3.5 years. The estimated probability of dian, >42 months) is significantly longer than that of the 92 patients
survival at 3 years is 50% (95% Cl, 42% to 58%). 30 to 59 years old (median, 25 months) or the 18 patients260 years
old (median, 1 month) (all comparisons, P < .001).

.OOl). The probability of survival at 3 years is estimated to

be 69% for those less than 30 years old, 39% for those 30 the L1 or L2 patients because of the small sample size. Their
to 59 years old, and 17% for those 2 6 0 years old (Fig 4). median WBC was 9,100/pL (range, 1 , 6 0 0 to 31,3OO/pL)
Survival is significantly longer for patients presenting with and none had a T or TMy immunophenotype or the Ph
a WBC less than 30,OOO/pL ( P = .001). Among patients chromosome. These patients have had a short median sur-
with high WBC counts, the median survivals are 24 months vival (6 months). Although 3 patients survived longer than
for those with 30,000 to 59,00O/pL, 14 months with 60,000 2 years, 2 have relapsed and 1of these subsequently received
to 99,OOO/pL, and 11 months with greater than lOO,OOO/pL, a BMT.
but these are not significantly different. Despite a high CR Survival is also influenced by immunophenotype: 38% at
rate, of the 25 patients with an initial WBC count greater 3 years for those with the expression of B-lineage antigens
than 100,OOO/pL, only 7 (27%) remain alive at last contact. compared with 69% for those with T-lineage antigen expres-
Patients who achieved a remission within 30 days are no sion ( P = .001; Fig 5 ) . The myeloid antigen-positive cases
more likely to be alive at 3 years than those who required have had an intermediate survival (55% at 3 years) but not
more than 30 days to achieve a CR (59% v 56%). statistically different from that of patients without myeloid
Survival is associated with morphology and immunophe- antigen expression within the same lymphoid lineage.
notype. The probability of survival at 3 years is estimated Similar to its impact on remission duration, the detection
to be 63% for L1 and 45% for L2 ( P = .03). The patients of the t(9;22) has a statistically significant adverse effect on
with L1 morphology had a median WBC count of 19,2001 survival. Only 16% (95% CI, 7% to 32%) of those with
pL (range, 800 to 401,OOO/pL); 39% had T or TMy immuno- either a Ph chromosome or a BCR-ABL rearrangement are
phenotypes and 20% were Ph positive. The L2 patients had estimated to survive for 3 years, compared with 62% of
a median WBC count of 17,2001pL (range, 700 to 328,5001 those who were negative by both genetic tests (P < .OOl).
pL); 21% had T or TMy immunophenotypes and 44% were In the multivariate analysis, after adjusting for age, WBC
Ph positive. The 8 patients with L3 morphology were consid- count retained statistical significance with respect to survival
ered as a separate FAB group but were not compared with (P< .00l). Among T or TMy patients, the WBC count did
not have prognostic significance for survival, but a mediasti-

1.0

0.8
P O
'l1-
R 0.8'
B 0.6 P
R
0
0.6-
i 0.4 O
R Buy (n-19)
O
N ; 0.4-

(1.2 O
N

0.0
0 1 2 3 4 3 6 7 8 9 1 0
"1'
0.0
I
Years 0 3 1 2 1 5 6

Years
Fig 3. The survival of all 197 eligible patients treated on CALGB
study8811 is comparedwith the survival of patients treated on three Fig 5. The survival for 39 patientswith a T or TMy immunophano-
previous CALQB studies: 7612 (185 patients), 8011 (308 patients), andWpe (median. >40 months) is significantly longer ( P = . M I ) than for
8513 (171 patients). 88 patients with a B or BMy immunophenotype (median, 21 months).
2032 LARSON ET AL

Table 5. Severe or Life-Threatening Toxicity a pulmonary embolism. Fatigue and malaise were common
Durina Treatment for ALL during L-asparaginase therapy; mental confusion was rare.
Induction Intensification Maintenance All patients were hospitalized during course I and many
Leukopenia (<2,000 yL) 98% 97% 75% patients required hospitalization during courses I1 and IV for
Thrombocytopenia (<50,OOO/fiL) 94 84 32 the treatment of fever while granulocytopenic. Otherwise,
Anemia (Hgb <8 g/dL) 65 84 26 hospitalization was seldom necessary. Increases in the levels
Hemorrhage 5 (1) 4 (2) 0 of hepatic transaminases were common and often required
infection 54 (7) 49 (4) 25 dose adjustments of maintenance chemotherapy. The treat-
Fever without infection 4 8 2 ment program was rigorous; 19 patients withdrew from the
Nausea/vomiting 8 17 8 prescribed therapy because of severe but less-than-life-
Stomatitis 7 9 7 threatening toxicity. These patients continued to be observed
Diarrhea 4 3 1
until relapse and death. Three patients underwent allogeneic
Hepatic 25 28 30
4
BMT in first remission; 2 of these have died. Two of the
Pulmonary 8 5 (1)
Cardiac 5 (1) 1 6 transplant patients had Ph-positive ALL; 1 reiapsed after
Genitourinary 8 (1) 2 1 BMT and has died and the other is alive and free of disease
CNS 6 13 6 more than 3 years after diagnosis. One patient underwent an
Peripheral nervous system 7 12 7 autologous BMT in first remission, relapsed 6 months later,
Skin 4 1 2 and died. A second patient, an 18-year-old manwithPh-
Allergy 0 1 1 positive ALL, underwent an autologous BMT in first remis-
The table liststhe frequencies (%) of grade 3 and 4 toxicities during sion followed by interferon therapyandwas alive atlast
each phaseof treatment using the CALGB Expanded Common Toxic- contact 14 months later. Two second cancers have been dis-
ity Criteria. The percentage of patients with lethal toxicity is shown covered, 1 renal and 1 breast carcinoma.
in parentheses. Patterns of failure. As described above, 28 patients
(14%) died while undergoing treatment on this study. Treat-
ment-related mortality was strongly related to increasing age
and was most often the result of infection.
nal mass was significantly associated with longer survival Twelve of the 13 patients who survived the induction
(P= .01). After controlling for age, WBC count, and medias- chemotherapy but failed to achieve a CR have died despite
tinal mass, FAB subtype retained prognostic significance additional therapy. One 19-year-old manhad Ph-positive
with respect to survival ( P = .001). If cytogenetic abnormali- ALL that was refractory to this protocol therapybut has
ties are accounted for first, then age, WBC count, and medi- remained free of disease for longer than 1 year after an
astinal mass continue to be significant. allogeneic BMT.
Toxicity. The major toxicities encountered during the in- Of 19 patients in remission who withdrew voluntarily or
duction and consolidation phases of this study were myelo- on their physician’s advice because of toxicity before com-
suppression and infection (Table 5). Seventeen patients (9%) pleting 2 years of the protocol therapy, 10 have experienced
died during the induction course, and most of these deaths a relapse of ALL and 8 of these have died. However, 7
were from infection with gram-negative bacteria or fungi. patients who did not complete the prescribed therapy remain
Nine of these patients were older than 60 years. There was alive with no evidence of disease. A11 7 have now been in
1 death from renal failure after tumor lysis syndrome during CCR longer than 3 years.
the induction phase. To date, 77 of the 167 patients (46%) who achieved a CR
Eleven patients (3 older than 60 years) died later, while have experienced a marrow and/or CNS relapse of ALL.
in remission and receiving consolidation or maintenance che- The minimum estimate of CNS relapse is 15% (25 patients).
motherapy. Six deaths were caused by infection, and l pa- Follow-up CSF examinations were not always reported, but
tient died after a colon perforation. Eleven patients were at least 7 patients have had concurrent relapses in the marrow
diagnosed to have Pneumocystis carinii pneumonia at some and the CNS. Only one patient, a 53-year-old man with Ph-
point during the 2 years of treatment; 1 of these patients positive ALL, had neurologic symptoms and CSF
died 3 weeks after the completion of all Chemotherapy.There lymphoblasts at diagnosis. Despite achieving a CR with in-
were 3 deaths from bleeding among patients in remission duction chemotherapy, intrathecal methotrexate, and cranial
who were receiving additional Chemotherapy: 2 hemorrhagic irradiation, he suffered a BM and CNS relapse 4 months
strokes and 1 gastrointestinal hemorrhage. after diagnosis and died 3 months later.
The L-asparaginase treatment was generally well toler- For 18 patients, CNS relapse preceded evidence of BM
ated. Several patients had local cutaneous reactions that did relapse by more than 1 month (range, 1 to 19 months). These
not recur when Erwinia L-asparaginase was substituted for latter patients had various features previously associated with
the initial Escherichia coli-derived drug. Treating physicians an increased risk of CNS leukemia: 5 had L3 morphology,
were required to evaluate the patient’s serum amylase level 5 had Ph-positive ALL, 3 had T-cell ALL, and 8 (including
before each L-asparaginase injection. Eight episodes of clini- 2 Ph-positive and 2 T-ALL cases) had greater than 30,000
cally significant pancreatitis were reported during courses I WBC/pL at diagnosis. Initially on this study, asymptomatic
and I1 during the 7 weeks of L-asparaginase therapy. One patients underwent their first lumbar puncture for spinal fluid
patient had inferior vena cava thrombosis and another had examination and intrathecal methotrexate at the beginning
NT INTENSIVE FOR ADULT ALL 2033

Table 6. The Impact of Unfavorable Pretreatment Characteristics onthe Survival of 197 Adults With ALL
Adverse Features

No. of Adverse No. of Age WBC FAB Ph or No Mediastinal % Estimated Survival at

Features Patients ~ 6 yr0 s30,WO/pL L3 t(4; 11) Mass 3 yr (95% Cl)

None 15 - - - - - 100 (77-100)

Exactly 1 42 0 6 0 0 36 74 (58-85)
1 Known 61 0 9 1 0 51 58 (42-72)
2 Known 55 7 55
31 6 11 25 (14-39)
3 Known 19 6 19
15 0 17 26 (12-49)
4 Known 5 5 5 1 4 5 0
Totals
18 197 66 8 32 166 50 (42-58)

The group reported to have one adverse feature is subdivided into 42 patients who had complete data regarding all five characteristics
(“exactly 1”) and 61 patients who had some missing data (mostly cytogenetics) but only one known adverse feature. Ph denotes Ph-positive
ALL detected by cytogenetics or by molecular assays.

of course HI, 3 months after diagnosis, when the BM was man (61 years old; W C 209,800/pL) had refractory disease
already in remission. Seven asymptomatic patients were dis- and died 2 months after diagnosis.
covered to have CSF lymphoblasts at that time. Isolated CSF High-risk patients. Groups of high-risk patients were
disease was also discovered in three patients who developed identified by the presence of one or more of the following
neurologic symptoms while receiving consolidation chemo- unfavorable characteristics: age 2 6 0 years, WBC count
therapy in course 11. Although most responded to intrathecal =30,00O/pL, FAB L3 morphology, absence of a mediastinal
chemotherapy and cranial irradiation and several maintained mass, and the presence of Ph or t(4; 11). Table 6 lists the
a subsequent CR for longer than l year, all eventually suf- numbers of patients with each of these features and their
fered a BM relapse or died from recurrent CNS leukemia. survival estimates. Fifteen patients (8%) had no unfavorable
After the protocol was amended so that intrathecal metho- features; 100% of these patients are estimated to be alive at
trexate was administered starting at the beginning of course 5 years (95% CI, 77% to 100%). Eight male and 7 female
I1 (see Table l), there were fewer cases of early CNS leuke- patients comprised this favorable group. Their median age
mia observed (2 patients). was 30 years (range, 16 to 39 years) and the median WBC
Six patients have had late isolated CNS relapses (without count was 10,700/pL (range, 1,400 to 24,9OO/pL). Five had
simultaneous BM relapse), occumng after more than 1 year L1 morphology and 10 had L2. Eighty-three percent had a
in CR. Four relapses occurred while receiving maintenance T or TMy phenotype. All had a mediastinal mass (by defini-
chemotherapy and 2 occurred 2 and 4 months after comple- tion), 5 had splenomegaly, and 3 had hepatomegaly.
tion of 2 years of therapy. All 6 have died, 3 after subsequent One hundred three patients (52%) had onlya single known
BM relapses. One 34-year-old man with T-cell ALL had an adverse feature; in 87, this was the absence of a mediastinal
isolated testicular relapse 5 months after completing 2 years mass. Fifty-nine percent of these patients were alive at last
of therapy. He achieved a second CR after local irradiation follow-up. For this analysis, missing values were imputed as
and received additional systemic chemotherapy; he suffered “no adverse feature,” which has undoubtedly caused some
a BM relapse 11 months later and died. underreporting of multiple adverse features, especially with
During the 24 months of treatment, 68 patients relapsed respect to cytogenetics. For example, only 49% of the pa-
or died in remission. The risk of failure was greater for tients listed with one unfavorable feature had cytogenetic
patients with Ph-positive or L3 ALL than for those with T- data available. Hence, we separated those patients with com-
lineage or B-lineage ALL butlacking the Ph or L3 character- plete information (“exactly one adverse feature”) from
istics. When considering only B-lineage (not including Ph- those with missing data and labeled the latter group as “one
positive or L3) versus T-lineage ALL cases, the risk of known adverse feature.” No patient is considered to be high
relapse was comparable throughout the 2 years of therapy. risk based solely on age. Patients known to have L3 morphol-
As yet, only 3 relapses have been observed among 50 pa- ogy or unfavorable cytogenetics appear to have multiple
tients in CR for longer than 3 years. other adverse features. There are no survivors among the 5
Of 30 patients identified to have the Ph chromosome or patients known to have four adverse features.
BCR-ABL rearrangement, 21 (70%) achieved a CR, but 16
of these have relapsed and died. Three others (39, 74, and DISCUSSION
80 years old) have died in remission, and 1 is alive with no One major goal of this study was to achieve initial cytore-
evidence of disease after a BMT. Only 1 patient (46 years duction as rapidly as possible, using five lymphocytotoxic
old) has remained on study and continues in first remission drugs during the induction course. The fraction of patients
for longer than 4 years. who subsequently achieved a CR (85%) was the highest yet
Patients with a t(4; 11) have been identified as a second observed in a CALGB trial and compares favorably with
cytogenetic group with a poor outcome.’’ Two such patients the outcomes reported from the best, large, single-institution
were known to be enrolled on this study. One woman (64 trials.2,6,7,9 The two preceding CALGB trials (8011 and
years old; WBC 392,2001pL) died during induction and 1 8513), enrolling unselected patients at the same centers with
 2034 LARSON ET AL

an equivalent age distribution, had shown CR rates of 64% more rapidly to chemotherapy have more durable remis-
71%,
and Other recent multicenter trials sions.16 Likewise, some studies of both childhood and adult
have also reported lower CRrates:74% for the German ALL have shown that patients obtaining a more rapid re-
ALL trial (GMALL-01). 75% for GMALL-02, 68% for the sponse of their leukemia to induction therapy have a higher
SWOG trial, and 64% for the ECOG trial, despite median CR rate and longer disease-free s ~ r v i v a l . ”However,
~~~ others
ages for the patients in each of these trials approximately 5 have found that time to achieve remission was not predictive
years younger than in the three CALGB trial^.^.'"^'' A recent of progn~sis.~’ In our study, we did not observe a statistically
Medical Research Council trial reported an 87% CR rate for significant difference in long-term outcome for the 27% of
patients with ALL, but that study included patients as young CR patients who required more than 30 days to enter remis-
as 14 years old and 40% of the patients were less than 20 sion. This finding might be explained by the fact that delay
years old.3’ in achieving CR is a function both of drug resistance of the
Among patients less than 30 years old, the CR rate of disease as well as myelotoxicity from the treatment.
94% in our study also exceeds the remission rates of 75% It would appear that the use of a more intensive remission
to 87% observed in other multicenter adult trials and ap- induction program has overcome the negative prognostic
proaches the 96% CR rate observed by the Children’s Cancer importance previously associated with the expression of my-
Group using the same induction schedule for high-risk child- eloid surface antigens or with any delay in achieving CR for
hood ALL.I3 Unfortunately, the response rate for patients longer than 30 days.'^'^,^^ When we carefully excluded pa-
more than 60 years old in our study was markedly affected tients with minimally differentiated AML (FAB type MO),
by the high early death rate. Thus, the question of whether we found no significant differences in the response rates,
the poor overall outcome of older adults with ALL is caused remission duration, or survival of ALL patients whohad
by inability to withstand treatment or to chemotherapy resis- coexpression of CD13 or CD33. In general, the course of
tance remains to be answered. However, once remission was these patients appeared to be determined by their lymphoid
achieved, the probability of continuing in CR at 3 years was lineage, ie, B or T cell.
not different between patients 30 to 59 years old and those The TMy immunophenotype may identify a distinct subset
older than 60 years, although the number of older patients with a favorable prognosis. We treated 8 such patients, all
was small. of whom were men. Their ages ranged from 17 to 46 years
Another objective of our trial was to provide the highest (median, 32 years). The median WBC count was 14,90O/pL
possible dose intensity of treatment during courses I and I1 (range, 1,OOO to 132,80O/pL). Three had splenomegaly, but
after initial rapid cytoreduction. For that reason, no delays only 1 had a mediastinal mass. Three had L1 morphology,
in treatment or dose reductions were permitted because of 3 had L2 morphology, and 2 were unclassified. None of the
pancytopenia in the absence of fever or obvious infection. 7 studied had the t(9;22) by cytogenetics or by molecular
Nevertheless, on average, patients required 4 months to com- assays. Seven (88%) achieved a CR; as yet, only l of these
plete the first 3 months of scheduled therapy. We cannot has relapsed and died. After 3 years, 86% remain in continu-
determine whether the apparent success of this treatment ous CR and 75% of the overall group remain alive.
program is the result of the initial rapid cytoreduction or the Fifteen percent of our patients had an enlarged mediasti-
dose intensity of the later therapy. num. Surprisingly, we found that this was predictive of a
Preliminary data from three randomized clinical trials sug- favorable outcome, with an influence on remission rate, dura-
gest that the concurrent use of filgrastim (granulocyte col- tion of remission, and survival in multivariate analyses. Most
ony-stimulating factor [G-CSF]) may improve the ability to previous studies of adult ALL have failed to identify medias-
deliver intensive chemotherapy more s a f e l ~ . ~The~ . ’ follow-
~ tinal enlargement as an important factor. A trend similar to
up periods for these trials are still short, and the full impact ours seemed apparent in an older German study, but it did
of the use of G-CSF during the treatment of adults with ALL not reach statistical significance.’ Furthermore, in our study,
remains to be determined. lymphadenopathy, another manifestation of leukemia mass,
Modem chemotherapy regimens for ALL have evolved had a favorable effect on remission rate and survival in a
empirically in such a fashion as to make it difficult to con- univariate but not multivariate analysis. This favorable effect
clude which part of the total therapy is most effective or, of mediastinal node enlargement may be caused in part by
alternatively, even deleterious. Few randomized clinical tri- the favorable effect of the T-lineage immunophenotype, be-
als have been performed. In this regard, it is important to cause 83% of the patients with an enlarged mediastinum had
note the results of the recent GIMEMA ALL-0288 trial, as T or TMy disease, or of L1 morphology (41% of these
yet published only in abstract form.35 Here, 541 patients patients). The median age of the patients with a mediastinal
between 12 and 65 years of age were randomly assigned to mass was 32 years (range, 16 to 54 years) and the median
remission induction treatment with daunorubicin, vincristine, WBC count was 24,lOOIpL (range, 1,400 to 475,000/&).
prednisone, and L-asparaginase with or without cydophos- None had the Ph chromosome.
phamide. Thus, one arm received the same five agents that Biologic characteristics of the disease continue to be pow-
weused in CALGB 881 1. However, those investigators erful determinants of response. The use of molecular meth-
found no difference in the CR rates (80%) on each m, ods to detect the BCR-ABL rearrangement will reliably de-
calling into question the added benefit of the cyclophospha- tect the approximately 30% of adults with ALL witha
mide. t(9;22).I8 Despite a nearly equivalent CR rate, the disease-
Patients with large-cell lymphoma whose disease responds free survival of these patients is clearly inferior to those
INTENSIVETREATMENT FOR ADULT ALL 2035

without this genetic mutation. Indeed, once this subgroup study supporting the usefulness of a dose-intense multicourse
was reliably identified and considered separately, the esti- 2-year therapy program for adults with ALL. The CR rate
mated 3-year survival of the remaining 3 1B-lineage patients is high, and for the younger than 30-year-old adult patients
improved to 48% (95% CI, 32% to 65%). This was not approaches that achieved in high-risk childhood ALL. The
significantly different (P = .13) from the survival of the 27 median survival has not been reached for younger patients;
similarly evaluated patients with T-lineage disease (62%; it is greater than 2 years even for those in the 30- to 59-
95% CI, 42% to 79%). As yet, no chemotherapy regimen year-old age group. Two subgroups have had particularly
has effectively cured patients with the Ph chromosome. In favorable outcomes: those with amediastinal mass and those
our study, only 16% (95% CI, 7% to 32%) are estimated to with a TMy immunophenotype. We have identified five pa-
survive at 3 years. The proportion of patients with the Ph tient characteristics that are related to adverse prognosis and
chromosome increases with age.40 Thus, this abnormality found that patients with multiple adverse features appear to
may be in part responsible for the poor prognosis of older have an increased risk. This observation requires validation
adults with ALL. Recent data suggest that approximately one using other data sets. Future efforts should be directed to-
third of Ph-positive patients can be cured using allogeneic ward designing innovative approaches to those patients with
BMT?’ At this time, allogeneic transplantation early in first the identified adverse prognostic factors, especially older age
CR should be considered the treatment of choice for Ph- and adverse cytogenetic/molecular biologic features.
positive patients of suitable age.
Burkitt-type (L3) B-ALL also had a poor outcome using ACKNOWLEDGMENT
this treatment program. Approximately 5% of adults with We thank Gwennelle Clark andAudrey Byrd for expert data
ALL have this subtype, which is characterized by the t(8; 14) coordination at the CALGB Data Management Center and Susan
or one of its variants, t(2;8) or t(8;22). Although 75% of Jarman for the preparation of the manuscript. The following main
our L3 patients achieved a CR, remissions were very short member CALGB institutions (principal investigator; N M grant sup-
(median, 3 months) and 5 of the 6 patients have relapsed. port) participated in this study, and we thank the many treating
Recent trials using short intensive chemotherapy programs physicians, cytogeneticists, research nurses, data managers, and pa-
with high doses of cyclophosphamide or ifosfamide and tients without whom it could not have been completed Bowman
high-dose methotrexate and cytarabine have produced dis- Gray School of Medicine (M. Robert Cooper, MD; CA 03927);
ease-free survival rates of 57% to 76% in children and adults Central Massachusetts Oncology Group-University of Massachusetts
Medical Center (Mary E. Costanza, MD); Dana-Farber Cancer Insti-
with L3 ALL or Burkitt’s lymph~ma:~-~~
tute (George P. Canellos, MD; CA 32291); Dartmouth Medical
Patients with FAB L2 morphology appear to do worse School-Noms Cotton Cancer Center (L. Herbert Maurer, MD; CA
than those with Ll, but it is not clear that L2 by itself can 04326); Duke University Medical Center (Jeffrey Crawford, MD;
be considered a poor-risk group. Indeed, in our multivariate CA 47577); Long Island Jewish Medical Center (Kanti R. Rai, MD;
analyses, of 10 patients whose only adverse feature would CA 11028); Massachusetts General Hospital (Philip Amrein, MD;
have been L2 morphology, all are alive at the most recent CA 12449);McGill Department of Oncology (Brian Leyland-Jones,
follow-up. Patients with L3 morphology or adverse cytoge- MD; CA 3 1809);Mount Sinai Hospital (James F. Holland, MD; CA
netics [Ph or t(4; l l)] appear to have other adverse features 04457); New York Hospital-Come11 Medical Center (Richard T.
and should be considered at high risk for treatment failure Silver, MD; CA 07968); North Shore University Hospital (Daniel
regardless of the known number of other adverse features. R. Budman, MD;CA 35279); Rhode Island Hospital (Louis A.
Complete data should be obtained at diagnosis to identify Leone, MD; CA 08025); Roswell Park Cancer Institute (Clara D.
Bloomfield, MD; CA 59518); SUNY Health Science Center at Syra-
the risk profile of individual patients with ALL to assist in
cuse (David B. Duggan, MD; CA 21060); University of Alabama,
treatment planning. Birmingham (George A. Omura, MD; CA 47545); University of
In 23% of our patients who have relapsed, lymphoblasts California at San Diego (Mark R. Green, MD; CA 11789); Univer-
were observed in the spinal fluid 1 to 19 months before a sity of Chicago (Nicholas J. Vogelzang, MD; CA 41287); University
BM relapse and another 9% had simultaneous CNS and BM of Iowa Hospitals (Gerald Clamon, MD, CA 47642), university of
relapses. Despite further treatment, all but 3 of these patients Maryland Cancer Center (Joseph Aisner, MD; CA 31983); Univer-
have died. In our amended protocol, the first specific CNS sity of Minnesota (Bruce A. Peterson, MD; CA 16450); University
treatment begins in course 11, after achieving a BM response. of MissourEllis Fischel Cancer Center (Michael C. Peny, MD; CA
It is likely that systemic cyclophosphamide, prednisone, and 12046); University of North Carolina at Chapel Hill (Thomas C.
Shea, MD, CA 47545); University of Tennessee, Memphis (Alvin
L-asparaginase provide some transarachnoid benefit during
M. Mauer, MD; CA 47555); Walter Reed A m y Medical Center
the induction course. However, the CSF appears to remain (Raymond B. Weiss, MD; CA 26806); and Barnes Hospital (Joanne
an important sanctuary site for ALL. Many of the CNS recur- Mortimer, MD; CA 47546).
rences occurred in patients with previously described high-
risk features for CNS involvement: Ph chromosome, L3 mor-
REFERENCES
phology, T-cell phenotype, or high initial WBC count. More
1. Hoelzer DF: Therapy of the newly diagnosed adult with acute
intensive systemic chemotherapy, such as high-dose metho-
lymphoblastic leukemia. Hematol Oncol Clin North Am 7:139, 1993
trexate or high-dose cytarabine, or higher radiation doses,
2. Kantarjian H M , Walters RS, Keating MJ, Smith TL, O’Brien
such as 3,000 cGy to the cranium and perhaps to the entire S, Estey EH, Huh YO, Spinolo J, Dicke K, Barlogie B, McCredie
spinal axis, might be appropriate for these poor-prognosis KB, Freireich EJ: Results of the vincristine, doxorubicin, and dexa-
patient^.^.^.^' methasone regimen in adults with standard- and high-risk acute
In summary, we report the results of a multi-institutional lymphocytic leukemia. J Clin Oncol 8:994, 1990
2036 LARSON ET AL

3. Hoelzer D, Thiel E, Ldffler, Buchner T, Ganser A, Heil G , acute lymphoblastic leukemia. Concordance among observers and
Koch P, Freund M, Diedrich H, Riihl H, Maschmeyer G, Lipp T, clinical correlations. Br J Haematol 47553, 1981
Nowrousian MR. Burkert M, Gerecke D, Pralle H, Mtiller U, Luns- 17. Bloomfield CD, Wurster-Hill D, Peng G, Le Beau M, Tantra-
cken Ch. Fulle H, Ho AD, KiichlerR, Busch F W , Schneider W, vahi R, Testa J, Davey FR, Ellison RR, Cuttner J, Schiffer C, Sobol
Gorg Ch, Emmerich B, Braumann D, Vaupel HA, von Paleske A, RE, Lunghofer B, Carey M, Mick R, Arthur D: Prognostic signifi-
Bartels H, Neiss A, Messerer D: Prognostic factors in a multicenter cance of the Philadelphia chromosome in adult acute lymphoblastic
study for treatment of acute lymphoblastic leukemia in adults. Blood leukemia, in Gale RP, Hoelzer D (eds): Acute Lymphoblastic Leuke-
71:123, 1988 mia. UCLA Symposium on Molecular and Cellular Biology, New
4. Gaynor J, Chapman D, Little C, McKenzie S, Miller W, An- Series, v01 108. New York, NY, Wiley-Liss, 1990, p 101
dreeff M, Arlin Z, Berman E, Kempin S, Gee T, Clarkson B: A 18. Westbrook CA, Hooberman AL. Spin0 C, Dodge RK, Larson
cause-specific hazard rate analysis of prognostic factors among 199 RA, Davey F, Wurster-Hill DH, Sobol RE, Schiffer C, Bloomfield
adults with acute lymphoblastic leukemia: The Memorial Hospital CD: Clinical significance of the BCR-ABL fusion gene in adult
experience since 1969. J Clin Oncol 61014, 1988 acute lymphoblastic leukemia: A Cancer and Leukemia Group B
5. Linker CA, Levitt L J , O’Donnell M, Forman SJ, Ries CA: Study (8762). Blood 80:2983, 1992
Treatment of adult acute lymphoblastic leukemia with intensive cy- 19. Sobol RE, Mick R, Royston 1, Davey FR, Ellison RR, New-
clical chemotherapy: A follow-up report. Blood 78:2814, 1991 manR, Cuttner J, Griffin JD, Collins H, Nelson DA, Bloomfield
6. Schauer P, Arlin ZA, Mertelsmann R, Cirrincione C, Friedman CD: Clinical importance of myeloid antigen expression in adult acute
A, Gee TS, Dowling M, Kempin S, Straus DJ, Koziner B, McKenzie lymphoblastic leukemia. N Engl J Med 316:1111, 1987
S, Thaler Tzvi H, Dufour P, Little C, Dellaquila C, Ellis S, Clarkson 20. Weiss RB, Vogelzang NJ, Peterson BA, Panasci LC, Carpen-
B: Treatment of acute lymphoblastic leukemia in adults: Results of ter JT, Gavignan M, Sartell K, Frei E, McIntyre OR: A successful
the L-IO and L-1OM protocols. J Clin Oncol 1:462, 1983 system for scientific data audits for clinical trials: A report from the
7. Clarkson B, Ellis S, Little C, Gee T, Arlin Z, Mertelsmann R, Cancer and Leukemia Group B. JAMA 270:459, 1993
Andreeff M, Kempin S, Cimncione C, Gaynor J: Acute lymphoblas- 21. Kaplan EL, Meier P Nonparametric estimation from incom-
tic leukemia in adults. Semin Oncol 12:160, 1985 plete observations. J Am Stat Assoc 53:457, 1958
8. Hussein KK, Dahlberg S , Head D, Waddell CC, DabichL, 22. Simon R, Lee YJ: Nonparametric confidence limits for sur-
Weick JK, Morrison F, Saiki JH, Metz E, Rivkin SE, Grever MR, vival probabilities and median survival time. Cancer Treat Rep
Boldt D, and the Southwest Oncology Group: Treatment of acute 66:37, 1982
lymphoblastic leukemia in adults with intensive induction, consoli- 23. Mantel N: Evaluation of survival data and two new rank order
dation, and maintenance chemotherapy. Blood 7357, 1989 statistics arising in its consideration. Cancer Chemother Rep 50: 163,
9. Radford JE Jr. Bums CP, Jones MP, Gingrich RD, Kemp JD, 1966
Edwards RW, McFadden DB, Dick FR. Wen B-C: Adult acute 24. Koziol JA, Reid N: On multiple comparisons among k sam-
lymphoblastic leukemia: Results of the Iowa HOP-L protocol. J Clin ples subject to unequal patterns of censorship. Comrnun Stat Theor
Oncol 758, 1989 Methods A6: 1149, 1977
10. Hoelzer D: Treatment of acute lymphoblastic leukemia. 25. Hosmer DW, Lemeshow S: Applied Logistic Regression.
Semin Hematol 31:1, 1994 New York, NY, Wiley, 1989
11. Gottlieb AJ, Weinberg V, Ellison RR, Henderson ES, Tere- 26. Cox D R Regression models and life tables. J R Stat SOCB
belo H, Rafla S, Cuttner J, Silver RT, Carey RW, Levy RN, Hutchin- 34:187, 1972
son JL, Raich P, Cooper MR, Wiernik P, Anderson JR, Holland JF: 27. SAS Technical Report P-229, SASISTAT Software: Changes
Efficacy of daunorubicin in the therapy of adult acute lymphocytic and Enhancements, Release 6.07. Cary, NC, SAS Institute, 1992, p
leukemia: A prospective randomized trial by Cancer and Leukemia 243 and 433
Group B. Blood 64:267, 1984 28. Pui CH: Acute luekemias withthe t(4;11)(q21;q23). Leuk
12. Ellison RR, Mick R, Cuttner J, Schiffer CA, Silver RT, Hen- Lymphoma 7:173, 1992
derson ES, Woliver T, Royston I, Davey FR, Glicksman AS, Bloom- 29. Cuttner J, Mick R, Budman DR, Mayer RI, Lee EJ, Hender-
field CD, Holland J F The effects of postinduction intensification son ES, Weiss RB, Paciucci PA, Sobol R, Davey F, Bloomfield C,
treatment with cytarabine and daunorubicin in adult acute lympho- Schiffer C: Phase 111 trial of brief intensive treatment of adult acute
cytic leukemia: A prospective randomized clinical trial by Cancer lymphocytic leukemia comparing daunorubicin and mitoxantrone:
and Leukemia Group B. J Clin Oncol 9:2002, 1991 A CALGB Study. Leukemia 5425, 1991
13. Steinherz PG, Gaynon P, Miller DR, Reaman G , Bleyer A, 30. Cassileth PA, Andersen W , Bennett JM, Hoagland HC,
Finklestein J, Evans RG, Meyers P, Steinherz LJ, Sather H, Ham- Mazza JJ, O’Connell MC, Paietta E, Wiernik P, for the Eastern
mond D: Improved disease-free survival of children with acute Cooperative Oncology Group: Adult acute lymphocytic leukemia:
lymphoblastic leukemia at high risk for early relapse with the New The Eastem Cooperative Oncology Group experience. Leukemia
York regimen -A new intensive therapy protocol: A report from 6: 178, 1992 (suppl 2)
the Children’s Cancer Study Group. J Clin Oncol 4:744, 1986 3 1. Durrant IJ, Richards SM: Results of Medical Research Coun-
14. Cheson BD, Cassileth PA, Head DR, Schiffer CA, Bennett cil trial UKALL IX in acute lymphoblastic leukaemia in adults:
JM, Bloomfield CD, Brunning R, Gale RP, Grever MR, Keating Report from the Medical Research Council Working Party on Adult
MJ, Sawitsky A, Stass S, Weinstein H, Woods WG: Report of the Leukaemia. Br J Haematol 85:84, 1993
National Cancer Institute-sponsored workshop on definitions of diag- 32. Larson RA, Linker CA, Dodge RK, George SL, Davey F R ,
nosis and response in acute myeloid leukemia. J Clin Oncol 8:813, Frankel SR, Powell BL, Schiffer CA: Granulocyte-colony stimulat-
1990 ing factor (filgrastim; G-CSF) reduces the time to neutrophil recov-
15. Bennett JM, Catovsky D, Daniel M-T, Flandrin G , Galton ery in adults with acute lymphoblastic leukemia receiving intensive
DAG, Gralnick HR, Sultan C: Proposals for the classification of remission induction chemotherapy: Cancer and Leukemia Group B
the acute leukemias. French-American-British (FAB) Co-operative Study 91 11. Proc Am SOCClin Oncol 13:305, 1994 (abstr 995)
Group. Br J Haematol 33:451, 1976 33. Ottmann OG, Hoelzer D, Gracien E, Kelly K, Ganser A,
16. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton Reutzel R, Lipp T, Busch F W , Schwonzen M, Wandt H, Heil G ,
DAG, Gralnick HR, Sultan C: The morphologic classification of Koch P, Hey11A, Meyer P, Bentz M, Peter S, Diedrich H, Kolbe
 INTENSIVETREATMENT FOR ADULT ALL 2037

K: Concomitant R-metHuG-CSF (filgrastim) and intensive chemora- distribution, BCR breakpoint and prognostic significance. Leukemia
diotherapy as induction treatment in adult ALL:A randomized 5196, 1991
multicenter phase I1 trial. Blood 82:193a, 1993 (abstr 757, suppl l ) 41. Barrett AJ, Horowitz MM, Ash RC, Atkinson K, Gale RP,
34. Ohno R, Tomonaga M, Kobayashi T, Kanamaru A, Shirakawa Goldman JM, Henslee-Downey PJ, Herzig RH, Speck B, Zwaan FE,
S , Masaoka T, Omine M, OhH, Nomura T, Sakai Y, Hirano M, Bortin MM: Bone marrow transplantation for Philadelphia chromo-
Yokomaku S , Nakayama S, Yoshida Y, Miura AB, Morishima Y, some-positive acute lymphoblastic leukemia. Blood 79:3067, 1992
DohyH, Niho Y, Hamajima N, Takaku F Effect of granulocyte 42. Reiter A, Schrappe M, Ludwig W-D, Lampert F, Harbott J,
colony-stimulating factor after intensive induction therapy in re- Henze G, Niemeyer CM, Gadner H, Muller-Weihrich S , Ritter J,
lapsed or refractory acute leukemia. N Engl J Med 323:871, 1990 Edenwald E, Riehm H: Favorable outcome of B-cell acute lympho-
35. Mandelli F, Annino L, Vegna ML, Camera A, Ciolli S, De- blastic leukemia in childhood: A report of three consecutive studies
plan0 W, Fabian0 F, Ferrara F, Ladogana S , Muti G, Peta A, Recchia of the BFM Group. Blood 802471, 1992
A, Sica S , Stasi R, Tabilio A, Visani G, Baccarani M for the GI- 43. Schwenn MR, Blattner SR, Lynch E, Weinstein HU: HiC-
MEMA group: GIMEMA ALL 0288: A multicentric study on adult COM: A 2-monthintensive chemotherapy regimen for children with
acute lymphoblastic leukemia. Preliminary results. Leukemia 6 182, stage I11 and IV Burkitt’s lymphoma and B-cell acute lymphoblastic
1992 (suppl 2) leukemia. J Clin Oncol 9:133, 1991
36. Armitage JO, Weisenburger DD, Hutchins M, Moravec DF, 44. P a w C, Philip T, Rodary C, Zucker J-M, Behrendt H, Gentet
Dowling M, Sorensen S, Mailliard J, Okerbloom J, Johnson PS, J-C, Lamagdre J-P, Otten J, Dufillot D, Pein F, Caillou B, Lemerle
Howe D, Bascom GK, Casey J, Linder J, Purtilo DT: Chemotherapy J: High survival rate in advanced-stage B-cell lymphomas and leuke-
mias without CNS involvement with a short intensive polychemo-
for diffuse large-cell lymphoma: Rapidly responding patients have
therapy: Results from the French Pediatric Oncology Society of a
more durable remissions. J Clin Oncol 4:160, 1986
randomized trial of 216 children. J Clin Oncol 9:123, 1991
37. Miller DR, Coccia PF, Bleyer WA, Lukens JN, Siege1 SE,
45. Fenaux P, Lai JL, Miaux 0, Zandecki M, Jouet JP, Bauters
Sather HN, Hammond GD: Early response to induction therapy as F: Burkitt cell acute leukaemia (L3 ALL) in adults: A report of 18
a predictor of disease-free survival and late recurrence of childhood
cases. Br J Haematol 71:371, 1989
acute lymphoblastic leukemia: A report from the Children’s Cancer 46. Cassileth PA, Andersen JW, Hoagland HC: High-dose cytara-
Study Group. J Clin Oncol 7:1807, 1989 bine therapy in adult acute lymphocytic leukemia, in Gale RP,
38. Hoelzer D: Therapy of acute lymphoblastic leukemia in Hoelzer D (4s): Acute Lymphoblastic Leukemia. New York, NY,
adults. Leukemia 6:132, 1992 (suppl 2) Liss, 1990, p 197
39. Drexler HG, Thiel E, Ludwig W-D: Review of the incidence 47. Wiernik PH, Dutcher JP, Paietta E, Gucalp R, Markus S ,
and clinical relevance of myeloid antigen-positive acute lymphoblas- Weinberg V, Azar C, Gar1 S, BensonL: Long-term follow-up of
tic leukemia. Leukemia 5:637, 1991 treatment and potential cure of adult acute lymphocytic leukemia
40. Secker-Walker LM, Craig JM, Hawkins JM, Hornrand AV: withMOAD:A non-anthracycline containing regimen. Leukemia
Philadelphia positive acute lymphoblastic leukemia in adults: Age 7:1236, 1993

View publication stats