Expert Interview
Treating Liver Metastases in Colorectal Cancer: An Expert Interview
With Michael Choti, MD



Medscape Hematology-Oncology. 2005; 8 (1): ©2005 Medscape
Editor's Note
In 2005, it is estimated that more than 104,000 cases of colon cancer
will be diagnosed in the United States alone[1] and that more than
56,000 patients will die of this disease.[1] The liver is the most
common site of colon cancer metastasis, and patients who undergo
resection of liver metastases can have prolonged survival. One
quarter to one third of patients who are able to undergo resection of
liver metastases will live 5 years or longer, and median survival
after resection is between 24 and 40 months.[2] Unfortunately, even
for patients with metastatic disease confined to the liver, not all
liver lesions are resectable, and even after successful resection,
microscopic disease elsewhere can lead to disease recurrence and
death. As novel systemic therapies that are proven to improve
survival in metastatic colorectal cancer emerge,[3] there is
increasing interest in treating patients who are potential candidates
for resection of metastases with preoperative and/or postoperative
chemotherapy.

Dr. Michael Choti and colleagues discussed recent advances and
emerging strategies for the multidisciplinary management of liver
metastases from colorectal cancer in an education session at the 2005
annual meeting of the American Society of Clinical Oncology (ASCO).
[4] In an interview with Medscape, Dr. Choti explains the promise and
limitations of adding chemotherapy to liver surgery and how he
currently approaches patient management in this area.

Medscape: Dr. Choti, this morning you and your colleagues discussed
preoperative, or neoadjuvant, and postoperative, or adjuvant,
chemotherapy for patients with colorectal cancer and potentially
resectable hepatic metastases. If metastatic colorectal cancer is a
systemic disease, what is the rationale for treating metastases to
the liver with local therapy such as surgery?

Dr. Choti: That is an excellent question. We don't know why
necessarily by resecting metastases we improve outcomes and why some
patients can actually be cured, but it does happen. So, number one,
we need to base the rationale for resection on the empiric data
showing that some patients can be cured with surgery alone. Whether
in those selected patients these are the only sites of disease or you
are cytoreducing a tumor down to a volume of disease in which, for
some reason, microscopic residual disease does not progress, we don't
know. This, in particular, is why combining effective systemic
chemotherapy with resection is a theoretically good strategy.

Medscape: In terms of the scope of this problem, what percentage of
patients with metastatic colorectal cancer have liver-only disease
and are potential candidates for resection?

Dr. Choti: It depends on which series is looked at. Approximately two
thirds of patients have liver as their first site of metastatic
disease, but many of these patients also have extrahepatic disease.
Roughly one third of patients who develop metastatic disease appear
to have disease confined to their liver on imaging studies.

Medscape: What percentage of patients with liver-only disease can
undergo resection with current strategies?

Dr. Choti: That varies with the aggressiveness of the surgeon.
Somewhere in the range of one third to one half of patients with
liver-only metastatic disease may be candidates for resection.

Medscape: This morning, you and your colleagues were discussing the
use of preoperative, or neoadjuvant, chemotherapy as one means to
improve outcomes for patients with liver metastases. Can you explain
the rationale behind neoadjuvant chemotherapy?

Dr. Choti: There are a few issues. One is that even in patients with
resectable disease, although we are making dramatic improvements in
long-term outcomes with surgery, many of these patients without any
additional therapy still experience recurrence. The strategy is to
combine effective systemic chemotherapy with surgery to improve
outcomes. Then it becomes primarily a sequencing question. In a
patient with resectable disease, is it preferable to give
chemotherapy upfront and then perform the surgery after assessing
response? Or is it better to resect the disease and then give
chemotherapy postoperatively when there is no residual measurable
disease? We don't know the answer to this question.

As we have discussed at this conference, there are pros and cons to
giving chemotherapy first. In a patient who initially has resectable
disease, the goal is not to give long and extensive treatments with
chemotherapy but to give therapy of a relatively limited duration,
then get the tumor out, and perhaps give more chemotherapy after
surgery. In some patients who have marginally resectable or
unresectable disease, we would need to give chemotherapy first and
consider the option of operating on them if and when their disease
becomes resectable.

In our multidisciplinary conferences, we assess the individual
patient and discuss the best paradigm for any individual patient's
management. First, we think in terms of whether a patient may fit
into a curative intent paradigm or is the goal primarily palliative?
If the patient's disease is unresectable initially but may become
resectable, then we think about them in a curative intent mode and
tailor the strategy with that goal in mind. We attempt to maximize
the response through choice of initial chemotherapy regimen to
improve the chance of resection, at the same time trying to limit
hepatotoxicity, which may make liver surgery more difficult.

If resection is unlikely and we need to consider the overall
treatment as palliative, then we sometimes use the chemotherapy
differently and save regimens to maximize median survival and quality
of life. Occasionally, in patients in whom we have initially thought
treatment was palliative, we have been dramatically surprised by
response to therapy and are able shift the strategy to a curative
intent and consider liver resection or ablation. Similarly, when a
patient who we initially believe has resectable disease progresses
during preoperative chemotherapy and develops unresectable disease,
palliation is the primary goal.

Medscape: Are there factors that you can use to predict who is likely
to benefit from a curative approach? Do we have markers of
sensitivity that can help determine who is likely to respond to
neoadjuvant chemotherapy?

Dr. Choti: There are a variety of factors, some of which are related
to chemotherapy and chemosensitivity. Many are related to factors
such as the number and location of the metastatic disease within the
liver and elsewhere. Other factors, such as disease-free interval,
the histology of the primary tumor, and nodal status of the primary
tumor, also have prognostic implications. We consider a response to
chemotherapy as another favorable prognostic factor. Perhaps, in the
future, we will have methods such as tumor expression profiling,
proteomics, and molecular genetics to more accurately predict which
patients will derive the greatest benefit from aggressive liver
surgery and chemotherapy.

Medscape: You mentioned that chemotherapy can be administered before
surgery, after surgery, or both before and after surgery. Do we know
or are there studies planned to clarify whether the timing of therapy
affects survival or other end points in treating colorectal cancer
patients with liver metastases?

Dr. Choti: We do not currently have studies to answer this question.
In rectal cancer, we know that preoperative therapy can control local
disease and allow us to limit the extent of resection. In metastatic
colorectal cancer, we don't have information on how timing of
chemotherapy affects outcomes, but we are interested in these
questions. There is an EORTC [European Organization for Research
Treatment of Cancer] trial that I believe has completed accrual in
which they are looking at preoperative and postoperative chemotherapy
vs no chemotherapy, but this may only answer the question of whether
chemotherapy helps. It does not compare preoperative vs postoperative
chemotherapy.

In patients with resectable disease, we do not know whether
neoadjuvant chemotherapy is beneficial. There are trials under
development that will attempt to answer this question. We hope to
have a trial like this under way in the near future. Right now there
are theoretical reasons to consider chemotherapy before or after
resection. I believe that the strategy at present should be to
increase awareness of the potential role of systemic therapy and the
understanding that the optimal way to manage patients with liver
metastases may be the combination of surgery and chemotherapy,
however it is given.

Medscape: This suggests that good communication between medical
oncologists and surgical oncologists is an important factor in the
care of these patients. To what extent do you think adequate
communication is occurring today?

Dr. Choti: I think this is a very important point. Multidisciplinary
management is important when considering options. Medical oncologists
need to have an understanding of when disease is resectable and when
surgery should be offered. Similarly, the surgical oncologist or
liver surgeon needs to have a comprehensive understanding of the
various chemotherapeutic options available, including the benefit and
potential toxicities. I believe that sessions such as the discussion
of this topic this morning at ASCO are aimed at improving
communication and understanding. Providers need to appreciate that,
although we cannot cure many patients, there should no longer be an
automatic nihilistic approach to all patients with metastatic
disease. We need to integrate chemotherapy and surgery in
aggressively selected patients with advanced colorectal cancer.

I believe surgeons are developing a real understanding of the role of
chemotherapy, and this needs to be brought into the mainstream of
surgical education.

Medscape: Is neoadjuvant chemotherapy for patients with liver
metastases from colorectal cancer ready for the clinic?

Dr. Choti: There are several theoretical advantages to preoperative,
or neoadjuvant, chemotherapy.

One advantage of giving chemotherapy upfront is that we can determine
whether the tumor is responsive or not before resection, as opposed
to simply giving chemotherapy empirically after resection.

Additionally, in selected cases there may be progression if there is
a biologically aggressive tumor. Identification of this subset
preoperatively can spare a patient surgical resection, which is
unlikely to be beneficial.

Another reason is that giving systemic therapy earlier theoretically
may improve outcomes as the chemotherapy is started earlier. Finally,
a response from preoperative chemotherapy may reduce the size of
liver metastases, allowing for less extensive surgery.

The theoretical disadvantages include the possibility that a patient
who has initially resectable disease could progress to an
unresectable situation. Although this is uncommon, losing the window
of opportunity to attempt curative therapy is unfortunate. However,
as I mentioned in my discussion this morning, patients whose tumors
progress during chemotherapy are probably not those who are going to
do well even if they undergo resection.

In addition, chemotherapy can result in hepatotoxicity, particularly
when administered for prolonged duration. This liver steatosis, if
severe, can in fact decrease rather than increase resectability and
may actually make the liver resection riskier. What type of
chemotherapy -- and for what duration -- is associated with increased
hepatotoxicity is not clear. We have even less information on newer
biologic agents such as bevacizumab, which may also affect wound
healing or liver regeneration. I think we don't have enough
information regarding these issues at this time, but many of us are
trying to shed light on this subject. Let's wait and see.

There are currently no clinical trials open asking these questions,
so we manage our patients selectively. In some we offer preoperative
chemotherapy, and in some we go right to surgery first, depending on
various factors. In patients who are chemo-naive with multiple
metastases and stage 4 disease at presentation, we may be more
inclined to offer chemotherapy upfront. If a patient is marginally
resectable, in which case a response would improve the outcome of
resection, we may be aggressive about giving chemotherapy upfront as
well. If a patient has comorbidities or there is concern about
chemotherapy causing hepatotoxicity in a patient who will require a
large resection, I may be more inclined to go to surgery first.

The other factor we consider is whether a patient has been exposed to
chemotherapy or not, either in the metastatic setting or in the
adjuvant setting. If a patient has received adjuvant chemotherapy for
his or her primary disease and disease recurred early, then we may be
less inclined to use neoadjuvant or adjuvant chemotherapy with
resection of liver metastases.

Medscape: What are the next steps we need to take in this field in
terms of research?

Dr. Choti: There are several important issues. With rapidly changing
options for chemotherapy, including the combination of cytotoxic
agents and biologic therapies, we need to define the optimal regimens
in various clinical scenarios. We need to consider how we integrate
the biologics with chemotherapy, what cytotoxic regimens we use, and,
as we've discussed, the sequencing of therapy needs to be worked out.

In terms of the biologics, there are additional questions. For
example, bevacizumab presents potential issues regarding
administration around the time of extensive surgery. We don't know if
this will have an impact on wound healing, bleeding, or liver
regeneration. These questions will be emerging in the context of the
multimodality approach to patients with liver metastases.

An additional important issue regarding neoadjuvant chemotherapy is
how do we manage the type of resection, particularly in patients who
had unresectable disease and then responded. Do you need to resect
the volume of tissue in which the initial disease resided or can you
only take the nidus of residual disease? Does a tumor respond
circumferentially? We don't really understand how much one needs to
resect. Currently, we try to resect all sites within the liver that
contained disease, but this needs to be more clearly defined.

Response to initial therapy raises additional questions. If the
patient responds, we typically use the same regimen after surgery. If
the patient's cancer progressed or even if it is stable disease,
should we offer a different chemotherapy regimen in the postoperative
adjuvant setting, or should we not offer any chemotherapy? This is
completely unknown and requires further investigation.

What is the final word? Well, this is an exciting and evolving time
for patients with advanced colorectal cancer and physicians treating
them, but we have many unanswered questions. Stay tuned.

References
Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer
J Clin. 2005;55:10-30.
Choti MA, Sitzmann JV, Tiburi MF, et al. Trends in long-term survival
following liver resection for hepatic colorectal metastases. Ann
Surg. 2002;235:759-766.
Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus
irinotecan, fluorouracil, and leucovorin for metastatic colorectal
cancer. N Engl J Med. 2004;350:2335-2342.
Choti MA. Treatment of liver metastases: what are the limits? what
are the goals? Management of hepatic colorectal metastases. Proc Soc
Am Clin Oncol. 2005:302-306. Education Session.


Funding Information

Supported by an independent educational grant from Sanofi-Aventis.



Michael Choti, MD , Associate Professor of Surgery and Oncology, The
Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins,
Baltimore, Maryland


Disclosure: Jeffrey Peppercorn, MD, MPH, has no disclosed no relevant
financial relationships.

Disclosure: Michael Choti, MD, has disclosed no relevant financial
relationships.