Transcript ASCO_2011_files/Ryan HIPEC Educ ASCO 2011

Colon Cancer
CS and HIPEC 2011
David P. Ryan, M.D.
Clinical Director, MGH Cancer Center
Tucker Gosnell Gastrointestinal Cancer Center
Associate Chief of Hematology/Oncology
2
3
The Agenda
• The argument against CS HIPEC
• The argument for CS HIPEC
• An example of a surgical series
• Review the two randomized studies in the field
• Final thoughts
4
CS and HIPEC
The Argument Against…
• Biologically flawed.
• Selection bias is inherent in all the surgical series.
• It’s toxic.
• The only way to answer the question is with a
randomized controlled trial.
5
CS and HIPEC
The Argument Against…
Any medical intervention should help people
1. Be Cured More Often
2. Live Longer
3. Feel Better
CS and HIPEC
The Argument For…
CS and HIPEC
The Argument For…
A.
Peritoneal Carcinomatosis is associated with a terrible
survival
CS and HIPEC
The Argument For…
A.
Peritoneal Carcinomatosis is associated with a terrible
survival
B.
Mechanical cytoreduction can cure or prolong survival
CS and HIPEC
The Argument For…
A.
Peritoneal Carcinomatosis is associated with a terrible
survival
B.
Mechanical cytoreduction can cure or prolong survival
C. Heating chemotherapy makes it more lethal to cancer
cells
CS and HIPEC
The Argument For…
A.
Peritoneal Carcinomatosis is associated with a terrible
survival
B.
Mechanical cytoreduction can cure or prolong survival
C. Heating chemotherapy makes it more lethal to cancer
cells
D. Intraperitoneal delivery allows higher doses and is more
lethal to cancer cells
CS and HIPEC
The Argument For…
A.
Peritoneal Carcinomatosis is associated with a terrible
survival
B.
Mechanical cytoreduction can cure or prolong survival
C. Heating chemotherapy makes it more lethal to cancer
cells
D. Intraperitoneal delivery allows higher doses and is more
lethal to cancer cells
E.
The Data: Surgical series suggest long term survival
A. Peritoneal Carcinomatosis from
CRC is Terrible
• Median survival of patients with peritoneal carcinomatosis
was 6-9 months in the 5FU era. (Chue DZ Cancer 1989)
• Only 1% of all metastatic patients regardless of peritoneal
carcinomatosis patients lived for 5 years (Dye Clin Col Ca
2009)
A. Peritoneal Carcinomatosis from
CRC is Terrible
• The Kohne model separates
patients into low, intermediate,
and poor risk based on
presence of 4 characteristics:
performance status, number of
involved sites, alkaline
phosphatase, and WBC count
A. Peritoneal Carcinomatosis from
CRC is Terrible…Not Necessarily
Lessons from N9741
Sanoff H K et al. JCO 2008;26:5721-5727
A. Peritoneal Carcinomatosis from
CRC is Terrible…Not Necessarily
Lessons from N9741
• Kohne risk associated
with OS
• PS associated with OS
Sanoff H K et al. JCO 2008;26:5721-5727
A. Peritoneal Carcinomatosis from
CRC is Terrible...Not Necessarily
Lessons from N9741
• Response to chemotherapy associated with 5
year survival
• 85/1682 patients survived for 5 years and
26% of them had a complete response to
chemotherapy
Sanoff H K et al. JCO 2008;26:5721-5727
A. Peritoneal Carcinomatosis from
CRC is Terrible...Not Necessarily
Lessons in the Modern Era
• Biology trumps everything else
• Patients with slow growing tumors do well
• Patients with tumors responsive to
chemotherapy do well
B. Mechanical Cytoreduction Can
Cure or Prolong Survival
• Approximately 20-30% of patients with isolated liver and
lung metastases will be cured
• The leap…
– Therefore, resection of isolated extrahepatic disease can also be
curable
– Therefore, mechanical cytoreduction can prolong survival in a
noncurative setting
B. Mechanical Cytoreduction Can
Cure or Prolong Survival
If someone has liver metastases
do we use this paradigm?
Standard
chemotherapy
Stage 4 colon
cancer
Standard chemotherapy +
mechanical cytoreduction
C. Heating Chemotherapy Makes it
More Lethal
C. Heating Chemotherapy Makes it
More Lethal
Klaver et al Ann Surg 2011
D. IP Chemo Allows Higher Doses
and Is More Lethal
Ovarian Cancer
Armstrong DK et al. N Engl J Med 2006;354:34-43.
D. IP Chemo Allows Higher Doses
and Is More Lethal
Ovarian Cancer
• Assumes either a dose
response problem or a
dose delivery problem
• Mitomycin and
cisplatin have no
activity when delivered
IV against colon ca
• No other use in
Armstrong DK et al. N Engl J Med 2006;354:34-43.
metastatic colorectal
cancer
E. CS + HIPEC
The Data: 5y Survival 12-75%
Yan et al Ann Surg Onc 2009
E. CS + HIPEC
The Data: Mortality 0-17%
Chua et al Ann Surg 2009
E. CS and HIPEC
The Netherlands Cancer Institute
• From 1995-2003, 117 patients with carcinomatosis from
CRC were treated at one center with CS and HIPEC
• Synchronous and metachronous patients less than age 71
and no liver/lung mets were eligible
• Complete macroscopic cytoreduction was attempted in
each patient
• HIPEC with mitomycin was used
• All patients received 5FU/LV or Irinotecan post CS and
HIPEC
Verwaal et al Ann Surg Onc 2005
E. CS and HIPEC
The Netherlands Cancer Institute
• 117 patients (97 colon, 15 appendix, 5 rectum)
– 59 had complete cytoreduction
– 44 had minimal tumor left behind (<2.5mm deposits)
– 14 had gross disease left behind
• 7 (6%) died of treatment related causes
• The median survival was 21.8 months
– 1year 75%
– 3year 28%
– 5year 19%
Verwaal et al Ann Surg Onc 2005
28
E. CS and HIPEC
The Netherlands Cancer Institute
• Extent of resection was the strongest predictor of survival
Verwaal et al Ann Surg Onc 2005
29
E. CS and HIPEC
The Netherlands Cancer Institute
• Criticisms
– Highly selected group of patients.
– Was there an ongoing selection bias?
– Which biology did better? Appendix?
– Which chemotherapy was administered?
– How many had chemotherapy prior to resection?
– How many had a response to chemotherapy?
Verwaal et al Ann Surg Onc 2005
30
E. CS and HIPEC
The Netherlands Cancer Institute
• Authors conclusions
• In other words…many are called but few are chosen!
Verwaal et al Ann Surg Onc 2005
E. CS and HIPEC
The Surgical Series
• Very few are controlled phase II studies.
• Patients come in all shapes and sizes…synchronous,
metachronous, different chemotherapy
• A strong trend over the years to only include those
patients in whom the surgeon feels that a complete
cytoreduction is possible
• In my own personal experience, recently the surgeons say
“come back in 6 months after FOLFOX/Bev”
• What we need is a randomized controlled trial!
32
CS HIPEC vs Standard Therapy
• The one randomized study
Verwaal V J et al. JCO 2003;21:3737-3743
CS HIPEC vs Standard Therapy
• Eligibility
– <72
– No evidence of metastatic disease on abd CT and cxray
– Normal liver and renal fxn
– First presentation of metastatic disease and no 5FU within last 12
months
Verwaal V J et al. JCO 2003;21:3737-3743
CS HIPEC vs Standard Therapy
.
Verwaal V J et al. JCO 2003;21:3737-3743
CS HIPEC vs Standard Therapy
Toxicity
.
• 7 (15%) developed a GI fistula
• 14 patients never started postop 5FU
• 4 patients (8%) died of surgery complications
Verwaal V J et al. JCO 2003;21:3737-3743
©2003 by American Society of Clinical Oncology
CS HIPEC vs Standard Therapy
.
Verwaal V J et al. JCO 2003;21:3737-3743
©2003 by American Society of Clinical Oncology
CS HIPEC vs Standard Therapy
.
©2003 by American Society of Clinical Oncology
CS HIPEC vs Standard Therapy
.
• Less burden of disease = better outcome
©2003 by American Society of Clinical Oncology
CS HIPEC vs Standard Therapy
Criticisms
• Prior to the era of multiagent chemotherapy…or was
it?
• What happens if we exclude the appendiceal
patients?
• Do we know that patients were randomized on the
ability of the most important stratification factor?
– Is the ability to cytoreduce prognostic, predictive, or both?
But we never proved it with liver
metastases???
Adam R, et al., Ann Surgery. 2004 Oct;240(4):644-57.
CS and HIPEC Criticisms
• Let’s assume that complete resection of disease if
possible is a good thing.
• We can’t tease out the independent effects of CS
and HIPEC
• Has anyone ever attempted to separate the two
components?
The other randomized trial
• Following CS, randomized patients to IP Chemo +
Standard 5FU/LV Chemo or Standard 5FU/LV
Chemo
Elias et al
IP Chemo versus Standard Chemo
following CS
Randomized Controlled Trials
• Represent the “gold standard” for evaluating new
therapies
– Eliminate biases of both patients and doctors
– Allow for accurate assessment of magnitude of effect in patients
meeting the eligibility requirements
– Law of Unintended Consequences
Randomized Controlled Trials
• Fat comes back but not where you expect it!
Randomized Controlled Trials
The Breast Cancer Experience
• Very good complete responses and even “cures” for HDT
and ASCT
• 5% mortality
• No need for randomized trials
Randomized Controlled Trials
The Breast Cancer Experience
Welch and Mogielnicki. BMJ
2002
Randomized Controlled Trials
The Breast Cancer Experience
Welch and Mogielnicki. BMJ 2002
Randomized Controlled Trials
The Breast Cancer Experience
“We conclude that initial conditions
matter; that conflicting values are
ubiquitous, pervade all stages of the
process, and permeate the judgment
of all parties to the discussion; and that
an institutional deficit exists in the
evaluation of procedures. Unlike the
evaluation of new drugs, which occurs
within a statutory framework,
administered by a federal agency,
governed by explicit rules, and
embedded in a culture and tradition,
the evaluation of procedures for which
there is no commercial sponsor is
much less organized.”
The Great Divide
Surgeons and Medical Oncologists
• Surgeons
– History of innovating
– Trained in the art of
surgical series
– Veni Vidi Vici
• Med Onc
– History of drug
development
– Trained to trust only in
phase III studies
– Nihilists
52
CS and HIPEC
My Conclusions
• Many are called but few are chosen
– All the surgical series are flawed by tremendous selection bias
CS and HIPEC
My Conclusions
• Many are called but few are chosen
• Toxicity is not trivial
– In an era when response to chemotherapy really matters, do we
really want to risk severe toxicity and even death?
– For something with a 3-5% mortality, isn’t it incumbent on us to
prove its worth?
CS and HIPEC
My Conclusions
• Many are called but few are chosen
• Toxicity is not trivial
• Complete surgical resection of metastatic
colorectal cancer may improve long term survival
and even cure the rare patient
CS and HIPEC
My Conclusions
• Many are called but few are chosen
• Toxicity is not trivial
• Complete surgical resection of metastatic
colorectal cancer may improve long term survival
and even cure the rare patient
• I highly doubt that HIPEC does anything of
positive consequence
CS and HIPEC
My Conclusions
We need 2 randomized trials for patients with
peritoneal carcinomatosis from colorectal cancer
1. Evaluating the role of cytoreductive surgery
2. Evaluating the role of HIPEC in those patients
who receive cytoreductive surgery