data from two separate studies
ratify earlier results showing that
following CHOP (cyclophosphamide
[Cytoxan, Neosar], doxorubicin
HCl, vincristine [Oncovin],
prednisone) with radiation improves
results for patients with early-
or limited-stage non-Hodgkin's
lymphoma (NHL), according to
data reported at the 43rd Annual
Meeting of the American Society of
Hematology (abstract 3023) .
The final results of E1484 showed
that radiotherapy was successful in
converting patients with limitedstage
diffuse aggressive NHL who
had achieved partial response following
treatment with CHOP to
complete response. In a trial conducted
by the Southwest Oncology
Group (SWOG), patients with early-
stage aggressive NHL receiving
CHOP plus radiotherapy continued
to realize survival advantages superior
to patients receiving CHOP
alone at greater doses.
New Results, Old System
In reporting the final results of
E1484, Sandra J. Horning, MD, of
Stanford University in California
noted that the study had "the advantage
of mature follow-up, but
the disadvantage of outdated classification,"
having been done before
the International Prognostic Index
(IPI) or the WHO and REAL classifications
Patient accrual and randomization
ran from October 1984 to September
1992. "The objectives of this
study," Dr. Horning explained,
"were to determine the complete response
rate and the toxicity of
CHOP for early-stage diffuse aggressive
lymphoma, to compare the effect
of involved field radiotherapy
after a CHOP-induced remission
with end points of duration of response,
survival, sites of relapse, and
toxicity, and lastly to determine the
ability of radiotherapy to convert
partial responders to complete responders."
Most patients in the study (82%)
had diffuse large-cell lymphoma. To
be eligible, patients had to be in stage
IE/II/IIE, or to have mediastinal, retroperitoneal,
or bulky disease, defined
as ≥ 10 cm if stage I.
The vast majority of patients had
a performance status of 0 to 1 and
slightly less than one-third had bulky
disease. A total of 81% of patients
had fewer than three disease sites,
according to the Ann Arbor classification.
Stage I disease was present
in 31% (14% stage I, 17% stage IE)
and the remainder had stage II disease.
Extranodal disease was present
in 47% of patients (and in this study
the spleen was considered an
extranodal site). The median age of
participants was 59 years.
Patients were initially stratified
based on performance status, tumor
mass, and number of disease sites.
"Randomization was up front," Dr.
Horning said. "Patients in both arms
of the study received a full eight cycles
of CHOP. Those in arm 1 who
achieved a complete response were
observed, whereas those in arm 2
received 30 Gy of involved field radiation.
Partial responders in either
arm received 40 Gy of radiation."
Among the 324 patients who completed
all eight cycles of chemotherapy,
the only four deaths were due to
CHOP toxicity. Among the worst
toxicities, 33% were grade 3 and 45%
were grade 4. Almost all toxicities
were restricted to neutropenia.
Conversions From Partial to
Among patients for whom response
data were available, "215, or
61%, achieved a complete remission
as defined in the protocol, with relatively
narrow confidence intervals,"
Dr. Horning reported. "Note that
28% of patients converted from partial
response to complete response
with the addition of radiation therapy,"
A smaller subset of patients actually
received consolidation treatment.
These were the 79 patients in arm 2
who achieved complete response and
went on to receive 30 Gy of involved
field radiation. In arm 1 there were
93 patients who achieved complete
response and were then observed.
Partial responders numbered 71.
Among those receiving consolidation
therapy, there was a 17% increase
in failure-free survival at 5
years and 15% at 15 years, Dr.
Horning reported. Time-to-progression
data showed only one relapse
after 6 years of treatment in either
arm of the study.
"I think it is notable that 78% of
the patients in the radiation therapy
arm are still in remission or estimated
to be in remission at 15 years," Dr.
Horning stated. Multiple regression
analysis showed performance status
of 2 to 4 and more than three involved
sites to be statistically significant adverse
factors for time to progression.
"Bulk, stage, extranodal disease, age,
and gender were not statistically significant,"
Dr. Horning said.
Curves for overall survival diverged,
came together, and again
separated somewhat at 15 years.
There were no statistical differences
in overall survival. "This study does
suffer from a small number of patients,"
Dr. Horning acknowledged.
"In conclusion, the complete response
rate in this study was 61%
after eight cycles of CHOP therapy.
We found that 28% of partial responders
converted to complete responders
with 40 Gy of involved field
radiotherapy. Overall, induction
CHOP of eight cycles was relatively
well tolerated," Dr. Horning said.
Important Difference Between
the Two Trials
Updated data from the SWOG
trial found that survival advantages
shown at 4.4 years median followup
for CHOP(3) (three cycles of
CHOP) plus radiation compared to
CHOP(8) (eight cycles of CHOP)
alone were essentially unchanged at
about 8.2 years. Overall survival
rates were 82% for CHOP(3) plus
radiation vs 74% for CHOP(8). Failure-
free survival rates were 76% for
CHOP(3) plus radiation vs 67% for
Patients eligible for the SWOG
trial had biopsy-proven intermediate-
or high-grade NHL, except lymphoblastic,
according to Thomas P.
Miller, MD, of Arizona Cancer Center
in Tucson. "The important difference
between this trial and the
one presented by Dr. Horning is that
nonbulky stage II and IIE, that is,
patients with a single mass measuring
10 cm or greater than one-third
chest diameter, were ineligible, presuming
three cycles of CHOP would
be inadequate therapy."