From Medscape Hematology-Oncology

Expert Interview
Improving Chemotherapy for Patients With Colorectal Cancer: An Expert
Interview With Dr. Mace Rothenberg
Posted 08/24/2006


Editor's Note:
Therapeutic options for patients with colorectal cancer, and
particularly for those with advanced disease, have expanded
substantially over the last 5 years. Although no new classes of
agents emerged in 2006, many investigators are focusing on improving
the efficacy and reducing the toxicities of current combination
regimens. In the adjuvant setting, researchers are attempting to
apply recent molecular findings to identify individuals for whom the
benefits of toxic chemotherapy are likely to outweigh the risks.
Following the 2006 annual meeting of the American Society of Clinical
Oncology (ASCO), Margie Miller, Editorial Director of Medscape
Hematology-Oncology, asked Mace Rothenberg, MD, Professor of Medicine
(Hematology/Oncology) and Ingram Professor of Cancer Research at
Vanderbilt-Ingram Cancer Center in Nashville, Tennessee, to provide
his perspectives on how recently presented findings might influence
patient care.

Medscape: What would you consider the highlights of this year's ASCO
meeting, with respect to treatment of colorectal cancer?

Dr. Rothenberg: OPTIMOX 2,[1] which was a follow up to OPTIMOX 1,[2]
looked at the question of giving patients who have responded to
chemotherapy and are stable a chemotherapy-free period. OPTIMOX 1
evaluated the feasibility of an oxaliplatin-free interval -- not a
chemotherapy-free period -- and found that patients who were treated
with this "stop-and-go" approach for oxaliplatin fared every bit as
well, in terms of overall survival, as those individuals on FOLFOX
[oxaliplatin/5-fluorouracil (5-FU)/leucovorin] who did not have an
oxaliplatin-free interval. So, OPTIMOX 2 took this question to the
next level and asked whether we can give a patient a chemotherapy-
free period. FOLFOX7 was given for 6 cycles to patients in both arms,
but then patients in the experimental arm, instead of continuing on 5-
FU/leucovorin maintenance (as in OPTIMOX 1), received no maintenance
until progression, when chemotherapy with FOLFOX7 was reintroduced.

The duration of disease control, which was defined as the time from
first response until the point at which the tumor became larger than
it had been at baseline, was equivalent in both arms. In OPTIMOX 1,
the duration of disease control was 12.9 months and in OPTIMOX 2, it
was 11.7 months; the median chemotherapy-free interval in OPTIMOX 2
was 4.5 months. So this trial demonstrates that the duration of
disease control is equivalent whether you continue to administer some
chemotherapy or stop all chemotherapy until the tumor progresses.

This is, in fact, what many of us have begun to do; that is, when
patients achieve maximum response and there's no further tumor
shrinkage, we offer time off treatment. With these data, as well as
prior information indicating that overall outcomes are not
compromised, we can feel more comfortable giving patients time off
treatment, allowing them to resolve toxicities, to have time spent
out of the hospital, out of the clinic, traveling, on vacation,
allowing their normal tissue time to recover. It is very much
appreciated by patients, and that's the bottom line.

The follow up to this study, the DREAM trial being conducted in
Europe, will evaluate maintenance therapy, not with chemotherapy, but
with targeted drugs, such as bevacizumab or cetuximab, alone.

The next important study was by Labianca and colleagues,[3] in which
they looked at intermittent vs continuous FOLFIRI [irinotecan/5-
FU/leucovorin] in patients with advanced colorectal cancer. The
question was whether the irinotecan-based FOLFIRI regimen could be
stopped and restarted in a similar way and with similar outcomes to
the experience with the oxaliplatin-based FOLFOX regimen. FOLFIRI was
administered as first-line treatment to patients with metastatic
colorectal cancer for a short 2-month period. If the patient had a
response, treatment was stopped for 2 months, and then after a 2-
month break, the FOLFIRI was restarted. Patients were randomized
either to this "stop-and-go" approach or to continuous FOLFIRI.
Response rates were a bit lower than what we have come to expect from
first-line chemotherapy, probably because neither of these treatment
arms included either bevacizumab or cetuximab. Progression-free
survival in both arms was nearly identical at about 6.5 months, and
overall survival was identical at 17.5 months in both arms. So, this
study demonstrated that even with irinotecan-based first-line
therapy, you can treat for a short period of time -- just 2 months --
and thereafter begin to introduce some treatment-free intervals.

I think 2 months is a bit short from an American perspective because
we like to continue chemotherapy for as long as we see continuing
tumor shrinkage. We only consider stopping once that response
stabilizes.

I think these studies are important because they give us some comfort
with regard to changing a practice that we had long worried about.
Many of us were trained to continue to give chemotherapy to patients
whose disease has stabilized because we thought it might be valuable
in suppressing the tumor. We worried that as soon as the chemotherapy
was stopped, the tumor would come roaring back. Now we know that that
is not the case.

Fuchs and colleagues presented the results of the BICC-C study,[4] a
phase 3 trial initially designed to look at irinotecan in combination
with 3 methods of administering fluoropyridines: bolus 5-FU 2 weeks
out of 3 (modified IFL); bolus + infusional 5-FU every 2 weeks
(FOLFIRI regimen); and daily capecitabine for 14 out of every 21 days
(CAPEIRI). There was subsequently a second randomization to celecoxib
vs placebo in an attempt to improve efficacy.

The original trial design for 1000 patients had to be modified after
about a year of accrual to incorporate the recently approved
angiogenesis inhibitor bevacizumab. The capecitabine arm (CAPEIRI)
was discontinued, and it became just a 2-arm study: FOLFIRI plus
bevacizumab vs modified IFL [bolus 5-FU and leucovorin] plus
bevacizumab, followed by a second randomization to celecoxib vs
placebo.

When the investigators looked at the first part of the study, the
progression-free survival for FOLFIRI was superior to that achieved
with either modified IFL or CAPEIRI, and when they corrected for
other prognostic factors, there continued to be a significant
advantage for FOLFIRI compared with the other 2 arms. Although
differences in overall survival did not reach statistical
significance, the trend once again was for FOLFIRI to be superior to
the other 2 arms. Overall, toxicities were similar, with a slight
decrease in the rate of febrile neutropenia in patients in the
FOLFIRI arm and a higher rate of hand-foot syndrome in those treated
with CAPEIRI. Of note, the incidence of myocardial infarctions or
acute vascular events including heart attack, stroke, and 60-day
mortality rates was slightly higher for the IFL regimen than for
either of the 2 other arms.

There were only very preliminary data for the second phase of the
study, which looked at the 2 treatment arms, both of which included
bevacizumab. It was very underpowered, but it seems that although
progression-free survival was similar, overall survival was trending
in favor of the FOLFIRI regimen compared with modified IFL. Median
survival was 18.7 months for patients treated on the modified IFL
plus bevacizumab arm compared with a median survival that had not
been reached for the FOLFIRI plus bevacizumab arm. The death hazard
ratio for modified IFL was 2.5 compared with FOLFIRI plus bevacizumab
(P = .01). Even with a very small sample size, infusional 5-FU seems
to have advantages over bolus administration when given in
conjunction with irinotecan.

These researchers concluded that first-line FOLFIRI was more
effective than the other 2 treatment arms in part 1 of this study, in
terms of progression-free survival and toxicity, with a trend towards
improved overall survival as well. In period 2, first-line FOLFIRI
plus bevacizumab was associated with significantly longer survival
than modified IFL. Finally, celecoxib neither improved efficacy nor
reduced chemotherapy-related toxicity. Even though this was a trial
that never reached its primary accrual goal and had to be modified on
a couple of occasions, it generated important data regarding FOLFIRI
as the preferred way of administering combination chemotherapy with
irinotecan in the first-line setting and provided additional evidence
against the use of modified IFL.

Another large and important study by Cassidy and colleagues[5]
examined the question of whether a compound known as xaliproden could
reduce the cumulative peripheral sensory neuropathy associated with
oxaliplatin. More than 600 patients who were to receive FOLFOX4 first-
line for metastatic colorectal cancer were randomized to also receive
either xaliproden orally 1 mg/day or a placebo. The primary end
points were reduction of the risk of grade 3/4 cumulative sensory
neuropathy and noninferiority in response rate. The concern about
response centers around whether something that protects against
toxicity also protects the tumor against the antitumor effects of
chemotherapy. Results showed that there was about a 40% reduction in
the risk of developing grade 3 neuropathy in those individuals who
received xaliproden. These findings were statistically significant.

The time to recovery in those who had developed neuropathy was the
same for both arms, so the key was a delay in the onset and a lower
rate of grade 3 peripheral neuropathy. The objective response rates
(43% to 45%) and survival (19 months) were virtually identical in the
2 arms. Adverse events that might be attributed to xaliproden were
not seen, so this seems to be a fairly safe drug with few significant
toxicities. The conclusion was that xaliproden is a potentially
useful drug in delaying the onset and reducing the severity of the
cumulative peripheral neuropathy associated with oxaliplatin.

I think that the implications of this study have to be assessed in
the context of some of the observations made with the "stop-and-go"
approaches we've just described. If you are treating a patient in the
metastatic setting and you are able to achieve a maximal response
very early [with OPTIMOX 2, it was only 3 months] -- which is well
before most patients will develop grade 3 peripheral neuropathy --
the role of a protective agent such as xaliproden may be somewhat
diminished. In the Cassidy study, patients were given FOLFOX4
continuously, which is why they had the rates of peripheral
neuropathy that were observed.

Howard Hochster presented results from the TREE-2 study,[6] which
built on the experience in TREE-1 in 2 ways. First, it added
bevacizumab to all 3 treatment arms [modified FOLFOX;
capecitabine/oxaliplatin (CAPEOX); bolus 5-FU/oxaliplatin/leucovorin
(bFOL)]. Second, it used a reduced dose of capecitabine because of
the high rate of toxicities seen in the CAPEOX arm of TREE-1. In TREE-
2, the overall response rate was about 50% in the FOLFOX plus
bevacizumab and CAPEOX plus bevacizumab arms and closer to 40% in the
bFOL plus bevacizumab arm. The time to tumor progression and overall
survival were similar in the CAPEOX and FOLFOX arms (10 months and 26
months, respectively) but were somewhat less for those who had
received the bFOL regimen (8 months and 20 months, respectively).

The investigators concluded that oxaliplatin plus bevacizumab, with
either infusional 5-FU or capecitabine, appeared to be active in this
setting, with an acceptable toxicity profile. They also noted that
the activity of bFOL plus bevacizumab was not sufficiently
encouraging to be pursued further. I think it was important to show
that the tolerability of the capecitabine regimen could be improved
by reducing the dose and that the therapeutic outcomes for CAPEOX
plus bevacizumab were quite similar to FOLFOX plus bevacizumab.
Results from a phase 3 trial comparing these 2 regimens head-to-head
are expected within the next 6 months.

Alan Venook presented the results of CALGB 80203, a phase 3 study[7]
of FOLFIRI or FOLFOX that added cetuximab or placebo as first-line
therapy and asked a very important question. Patients were first
randomized to either FOLFOX or FOLFIRI, and then to either cetuximab
or placebo. The study was originally designed to accrue 2200 patients
with metastatic colorectal cancer but was closed by the Data and
Safety Monitoring Board after enrolling only 238 patients, primarily
because of the approval of bevacizumab in this setting. It was felt
that patients who were eligible should be receiving bevacizumab as
part of front-line therapy.

The primary end point of the study was to determine whether the
overall survival for patients with EGFR-expressing tumors treated
with cetuximab would be better than the survival in patients treated
with chemotherapy alone. Although the low accrual and short median
follow-up of the study prevented definitive conclusions from being
drawn at the time of the presentation, data on response rates and
progression-free survival were somewhat inconsistent. While the
addition of cetuximab to first-line chemotherapy appeared to improve
response rate -- 52% in those who received cetuximab vs 38% in those
who did not -- this difference did not carry over to progression-free
survival, where those who received chemotherapy plus cetuximab had a
median progression-free survival of 8.5 months and those who received
chemotherapy alone had a median progression-free survival of 9.4
months. It is possible that the impact of cetuximab may vary with the
specific chemotherapy regimen. For instance, the response rates were
about 40% for the FOLFIRI alone, FOLFIRI/cetuximab, and FOLFOX-alone
arms, but approximately 60% for the FOLFOX plus cetuximab arm.
However, the number of patients enrolled in this trial fell far short
of its goal, and any conclusions regarding comparative efficacy have
to be considered tentative, at best.

This study has now been replaced by an ongoing intergroup trial,
C80405, that is randomizing patients who receive either FOLFOX or
FOLFIRI to cetuximab, bevacizumab, or the combination. It is planned
to include more than 2200 patients.

Medscape: Moving from the metastatic setting to the adjuvant setting,
how should clinicians identify those patients with high-risk stage II
disease who are most likely to benefit from adjuvant therapy?

Dr. Rothenberg: Pooled analyses of stage II patients yield
conflicting results. According to the NSABP, for instance, the
magnitude of the impact of chemotherapy in the adjuvant setting is
around 30%, regardless of whether the patient has stage II or stage
III disease. I believe that this is correct. Unfortunately, because
the survival of patients with stage II disease is so much better than
for those with stage III, trials have failed to show a statistically
significant advantage for stage II patients. The question boils down
to: can we justify giving all patients chemotherapy when we know that
the majority of those with stage II disease are not destined to recur?

A trial that would randomize unselected patients with stage II
disease to chemotherapy vs no chemotherapy would require many
thousands of subjects. Moreover, although FOLFOX turns out to be a
more effective adjuvant regimen than 5-FU/leucovorin,[8,9] the use of
oxaliplatin introduces a toxicity, mainly neuropathy, that may be
long lasting and possibly permanent in some patients.

In an attempt to address these issues, an ongoing trial, E5202,[10]
is randomizing patients with locally advanced colon cancer to
chemotherapy or no chemotherapy on the basis of molecular
characterizations -- microsatellite instability and 18q loss of
heterozygosity -- that, in retrospective analyses, were associated
with poor prognosis.[11] Those considered to be "low risk" are
assigned to observation (ie, no chemotherapy). Those deemed to
be "high risk" are randomized to either FOLFOX alone or FOLFOX plus
bevacizumab. This trial opened in August 2005 and has accrued 121 of
a planned 3600 patients.

Ideally, we would randomize all stage II patients to either
chemotherapy or no chemotherapy, but the feeling was that it was
unethical to withhold treatment from those in whom bad outcome is
predicted. The study question is slightly different, therefore, in
that we will discover whether high-risk patients benefit from the
addition of bevacizumab.

Molecular characterization has finally emerged as a means of
selecting and assigning patients to treatment, and we are at least
taking a small step forward in determining what might be the best
approach for patients with stage II colon cancer.

Medscape: Looking ahead 12 to 18 months, what do you think are likely
to be the next significant therapeutic advances in colorectal cancer?

Dr. Rothenberg: We can't expect to see the same pace of progress in
advanced colorectal cancer that we've seen over the past 5 years. We
have been very fortunate to have a number of new drugs demonstrate
significant benefits in this disease, leading to a doubling of the
overall survival compared with 10 years ago. However, current trials
are really focused on refining those available treatments, learning
about ways to reduce toxicity, reduce cost, increase patient
acceptance, and increase tolerability. We want to determine whether
we can combine these drugs in an acceptable fashion.

A small randomized trial looked at first-line treatment with all
active cytotoxic agents together (FOLFOXIRI)[12] vs FOLFIRI, and the
response rates -- 66% for FOLFOXIRI, 41% for FOLFIRI -- were
significantly different. The percentage of patients who could
subsequently undergo potentially curative resections, 15% for the
FOLFOXIRI, 6% for FOLFIRI, were again significantly different.
Progression-free survival was improved for FOLFOXIRI compared with
FOLFIRI (9.8 months vs 6.9 months), and overall survival also favored
FOLFOXIRI -- 22.6 months vs 16.7 months. This small trial
demonstrated for the first time that combining all 3 active
chemotherapy drugs may be associated with certain advantages in terms
of efficacy. Not surprisingly, though, the rate of some toxicities
such as grade 3/4 neutropenia (50%) and febrile neutropenia, was also
increased. The incidence of grade 3/4 diarrhea was doubled, the
incidence of grade 3/4 vomiting was doubled, neurotoxicity was
higher -- this regimen was not without significant costs in terms of
patient toxicity. In today's world, where all eligible patients
receive biologicals as well as chemotherapy, one has to think about
the implications of these ever-expanding regimens.

For instance, if we added bevacizumab to FOLFOXIRI, would we achieve
further improvements? If so, would there be even greater efficacy by
combining FOLFOXIRI with bevacizumab plus cetuximab? In the past, we
knew that in most solid tumors, once we got much beyond 2 or 3
cytotoxic drugs, we reached the point of diminishing returns. Four
drugs are not better 3, and in many diseases, 3 are not better than
2. Nevertheless, these rules may not apply to biologicals, and trials
are underway to address this question in colorectal cancer.

Medscape: Do you have any "take-home" messages for practitioners
about treating patients with colorectal cancer?

Dr. Rothenberg: I think that we have to keep in mind that the
toxicities of these regimens are not trivial and that we have to
carefully monitor patients for diarrhea and neutropenia, especially
when they occur together. For patients with grade 3 neuropathy
associated with oxaliplatin or any vascular event associated with
bevacizumab, that agent should be stopped or switched. Patients on
capecitabine require close monitoring to make sure that the dose is
interrupted or adjusted at the earliest signs of hand-foot syndrome
or diarrhea. In fact, a poster was presented at this meeting by
Haller and colleagues[13] about the increased toxicity of
fluoropyridines, either capecitabine or 5-FU/leucovorin, in American
patients compared with patients in Europe or the Far East. We have to
be careful about managing these patients with more active, but also
potentially more toxic, therapies.

References
Maindrault-Goebel F, Lledo G, Chibaudel B, et al. OPTIMOX2, a large
randomized phase II study of maintenance therapy or chemotherapy-free
intervals (CFI) after FOLFOX in patients with metastatic colorectal
cancer (MRC). AGERCOR study. Proc Am Soc Clin Oncol. 2006;24:147s.
Abstract 3504.
Tournigand C, Cervantes A, Figer A, et al. OPTIMOX1: a randomized
study of FOLFOX4 or FOLFOX7 with oxaliplatin in a stop and go fashion
in advanced colorectal cancer -- a GERCOR study. J Clin Oncol.
2006;24:394-400. Abstract
Labianca R, Floriani I, Cortesi E, et al. Alternating versus
continuous "FOLFIRI" in advanced colorectal cancer (ACC): a
randomized "GISCAD" trial. Proc Am Soc Clin Oncol. 2006;24:147s.
Abstract 3505.
Fuchs C, Marshall J, Mitchell E, et al. A randomized trial of first-
line irinotecan/fluoropyrimidine combinations with or without
celecoxib in metastatic colorectal cancer (BICC-C). Proc Am Soc Clin
Oncol. 2006;24:147s. Abstract 3506.
Cassidy J, Bjarnason GA, Hickish T, et al. Randomized double blind
(DB) placebo (Plcb) controlled phase III study assessing the efficacy
of xaliproden (X) in reducing the cumulative peripheral sensory
neuropathy (PSN) induced by the oxaliplatin (Ox) and 5-FU/LV
combination (FOLFOX4) in first-line treatment of patients (pts) with
metastatic colorectal cancer (MCRC). Proc Am Soc Clin Oncol.
2006;24:147s. Abstract 3507.
Hochster HS, Hart LL, Ramanathan RK, et al. Safety and efficacy of
oxaliplatin/fluoropyrimidine regimens with or without bevacizumab as
first-line treatment of metastatic colorectal cancer (mCRC): final
analysis of the TREE-Study. Proc Am Soc Clin Oncol. 2006;24:148s.
Abstract 3510.
Venook A, Niedzwiecki D, Hollis D, et al. Phase III study of
irinotecan/5FU/LV (FOLFIRI) or oxaliplatin/5FU/LV (FOLFOX){±}
cetuximab for patients (pts) with untreated metastatic adenocarcinoma
of the colon or rectum (MCRC): CALGB 80203 preliminary results. Proc
Am Soc Clin Oncol. 2006;24:148s. Abstract 3509.
Andre T, Boni C, Mounedji-Boudiaf L, et al. Oxaliplatin, fluorouracil
and leucovorin as adjuvant treatment for colon cancer. N Engl J Med.
2004;350:2343-2351. Abstract
De Gramont A, Boni C, Navarro M, et al. Oxaliplatin/5FU/LV in the
adjuvant treatment of stage II and stage III colon cancer. Efficacy
results with a median follow-up of 4 years. Proc Am Soc Clin Oncol.
2005;23:246s. Abstract 3501.
Benson AB. Present and future role of prognostic and predictive
markers for patients with colorectal cancer. ASCO 2006 Educational
Book, pp 187-190.
Watanabe T, Wu TT, Catalano PJ, et al. Molecular predictors of
survival after adjuvant chemotherapy for colon cancer. N Engl J Med.
2001;344:1196-1206. Abstract
Falcone A, Masi G, Murr R, et al. Biweekly irinotecan, oxaliplatin,
and infusional 5FU/LV (FOLFOXIRI) versus FOLFIRI as first-line
treatment of metastatic colorectal cancer (MCRC): Results of a
randomized, phase III trial by the Gruppo Oncologico Nord Ovest
(GONO). Program and abstracts of the 2006 Gastrointestinal Cancers
Symposium; January 26-28, 2006; San Francisco, California. Abstract
199.
Haller DG, Cassidy J, Clarke S, et al. Tolerability of
fluoropyrimidines appears to differ by region. Proc Am Soc Clin
Oncol. 2006;24:149s. Abstract 3514.


Related Links
Resource Centers
Colorectal Cancer


Funding Information

Supported by an independent educational grant from sanofi-aventis.




Mace Rothenberg, MD, Professor of Medicine, Vanderbilt University
Medical School, Nashville, Tennessee; Attending Physician, Vanderbilt
University Medical Center, Nashville, Tennessee


Disclosure: Margie Miller has disclosed no relevant financial
relationships.

Disclosure: Mace Rothenberg, MD, has disclosed that he has received
grants for clinical research from and has served as an adviser or
consultant to Pfizer, sanofi-aventis, Roche, and ImClone. Dr.
Rothenberg has also disclosed that he has received grants for
educational activities from Pfizer and that he has served as an
advisor or consultant to Genentech.


Medscape Hematology-Oncology. 2006;9(2) ©2006 Medscape