вход по аккаунту



код для вставкиСкачать
A Festschrift Honoring the 90th Birthday on May 9, 1997, of
Jonathan Evans Rhoads, M.D., Editor Emeritus of Cancer
Cancer special section
Systemic Chemotherapy for Gastric Carcinoma
followed by Postoperative Intraperitoneal Therapy
A Final Report
Peter Crookes, M.D.1
Cynthia Gail Leichman, M.D.2
Lawrence Leichman, M.D.2
Matthew Tan, M.D.3
Loren Laine, M.D.2
Steven Stain, M.D.1
Joaquina Baranda, M.D.2
Yolee Casagrande, R.N.1
Susan Groshen, Ph.D.4
Howard Silberman, M.D.1
BACKGROUND. Because only approximately 50% of gastric carcinomas are resectable for cure, the authors hypothesized that effective systemic preoperative (neoadjuvant) chemotherapy, aimed at decreasing the size and extent of the primary
Department of Surgery, University of Southern
California School of Medicine, the Los Angeles
County / University of Southern California
Medical Center, and the University of Southern
California-Kenneth Norris Cancer Hospital, Los
Angeles, California.
Department of Medicine, University of Southern California School of Medicine, the Los
Angeles County / University of Southern California Medical Center, and the University of
Southern California-Kenneth Norris Cancer Hospital, Los Angeles, California.
Department of Radiology, University of Southern California School of Medicine, the Los
Angeles County / University of Southern California Medical Center, and the University of
Southern California-Kenneth Norris Cancer Hospital, Los Angeles, California.
Department of Preventive Medicine, University
of Southern California School of Medicine, the
Los Angeles County / University of Southern
California Medical Center, and the University of
Southern California-Kenneth Norris Cancer Hospital, Los Angeles, California.
Supported in part by NIH CA-RO1 561119-02
and a grant from Hoffman-LaRoche, Incorporated, Nutley, New Jersey.
Address for reprints: Lawrence Leichman, M.D.,
USC Department of Medicine, USC/Norris Cancer Hospital and Research Institute, 1441 Eastlake Avenue, Los Angeles, CA 90033.
Received December 18, 1996; revision received
February 18, 1997; accepted February 18, 1997.
tumor and eradicating distant microscopic disease, may increase the rate of resectability and have a greater impact on survival than postoperative (adjuvant) treatment alone. In addition, because the peritoneal cavity is the most common site
of first recurrence after successful gastric cancer resection, intraperitoneal (IP)
chemotherapy seemed a logical choice for postoperative (adjuvant) treatment.
METHODS. Fifty-nine patients with invasive primary gastric adenocarcinoma who
were deemed resectable for cure entered a clinical trial that called for 2 cycles of
protracted infusion 5-fluorouracil with weekly leucovorin and cisplatin chemotherapy followed by surgery. Approximately 3–4 weeks after potentially curative surgery, patients were scheduled to receive two cycles of IP 5-fluoro-2*deoxyuridine
and cisplatin.
RESULTS. Of the 59 patients studied, 58 (98%) received both cycles of systemic
chemotherapy. Fifty-six patients (95%) underwent surgery: 40 patients (71%) had
resections intended to cure for Stage 0-IIIB disease, 15 patients (27%) had palliative
surgery for Stage IV gastric carcinoma, and one patient died intraoperatively without being staged. Two patients refused surgery, and the remaining patient died of
progressive disease prior to surgery. Thirty-one of the 40 patients who underwent
curative surgery completed both cycles of postoperative IP therapy; 4 patients
received only 1 cycle. Three patients (5%) died secondary to treatment complications. There were two operative deaths, and one patient died of peritonitis associated with Grade 4 granulocytopenia. Nine of the 40 patients (23%) whose carcinomas were resected for cure had recurrent carcinoma. With a median follow-up
period now exceeding 45 months, the calculated median survival for the 59 patients
entered into the trial is ú4 years.
CONCLUSIONS. This program of preoperative systemic and postoperative IP chemotherapy has been found to be safe and appears to decrease gastric carcinoma
recurrence rates and increase survival compared with historic controls. Cancer
1997;79:1767–75. q 1997 American Cancer Society.
KEYWORDS: gastric carcinoma, adjuvant chemotherapy, preoperative chemotherapy, intraperitoneal chemotherapy.
astric carcinoma is a significant cause of death in the U.S., especially in areas populated by first-generation Hispanic and Asian
immigrants, among whom there is an inordinately high incidence of
this disease. The growing population of these ethnic groups served by
the hospitals of the University of Southern California (USC) Medical
Center has tended to focus the authors’ attention on adenocarcinoma
of the stomach. In fact, gastric carcinoma represents the leading cause
q 1997 American Cancer Society
/ 7b55$$1073
04-02-97 14:46:42
W: Cancer
CANCER May 1, 1997 / Volume 79 / Number 9
of death from gastrointestinal malignancy at the Los
Angeles County-USC Medical Center.1 Although the
ethnic background of gastric carcinoma patients seen
at USC differs from most other institutions, the advanced stages of disease and poor survival data are
similar to other gastric carcinoma cohorts treated elsewhere in the U.S. and Europe. In fact, Diehl et al.2
reported that only 28% of 2867 patients with adenocarcinoma of the stomach treated at 5 centers in the U.S.
had a stage of disease amenable to curative surgery,
and no center reported a 5-year survival rate exceeding
15%. In addition, among patients who have undergone
surgery intended to cure, approximately 50% develop
initial recurrence in the peritoneal cavity3,4 regardless
of the anatomic site of the primary tumor within the
More sophisticated preoperative staging with
computerized tomography (CT), endoscopic ultrasound, and diagnostic laparoscopy may produce a better rate of postoperative survival by limiting surgical
exploration to patients with a higher probability of
cure, but improved overall survival of patients with the
extent of disease currently prevalent awaits improved
treatment modalities, which theoretically may include
more effective radical surgery, preoperative therapy to
downstage the primary tumor and treat disseminated
microscopic disease and postoperative therapy to prevent recurrence and address the major site of initial
failure, the peritoneal cavity.
Because neither preoperative neoadjuvant chemotherapy alone5,6 nor postoperative adjuvant therapy alone7,8 has yet resulted in improved survival for
most patients with gastric carcinoma, the authors developed a program that combined preoperative systemic chemotherapy comprised of two cycles of protracted infusion (PI) 5-fluorouracil (5-FU) with weekly
leucovorin and cisplatin with a postoperative adjuvant
component providing two cycles of intraperitoneal
(IP) 5-fluoro-2*-deoxyuridine (FUdR) and cisplatin.
The agents selected for preoperative combination
chemotherapy were chosen because they cause minimal bone marrow toxicity prior to surgery when administered with the schedule employed. In addition,
neither cytotoxic agent, 5-FU or cisplatin, interferes
with recovery from an extensive surgical procedure,
and, most important, each drug is known to be relatively active against gastric carcinoma.9 – 11 The IP therapy included FUdR and cisplatin. The dose of IP FUdR
chosen results in peritoneal fluid drug levels two to
three logs higher than the plasma counterparts.12
Moreover, IP FUdR is well tolerated and not associated
with the high risk of chemical peritonitis observed with
IP 5-FU.13 IP cisplatin administration also confers a
substantial increase (approximately one log) in peritoneal fluid concentration compared with plasma.14 The
/ 7b55$$1073
04-02-97 14:46:42
potential benefit of this increased peritoneal concentration is suggested by Markman et al.,15 who found
that patients with small peritoneal metastases from
ovarian carcinoma that were refractory to systemic cisplatin had a 42% response rate when the drug was
administered by the IP route. Because the clinical and
hematologic toxicities of IP FUdR and cisplatin do not
overlap, it was assumed that these agents could be
safely delivered in combination to patients who had
undergone potentially curative gastric resections.
The first publication presenting the early results
of this protocol described 38 patients with primary
gastric carcinoma after a median follow-up period of
19 months. The median survival exceeded 17
Eligibility Criteria
Patients aged at least 18 years with histologically
proven gastric adenocarcinoma not previously treated
were considered for eligibility. Tumors involving the
cardia were included if they appeared to be arising
from the upper stomach or fundus rather than the
esophagus. Initial evaluation included a complete history and physical examination, chest X-ray, upper gastrointestinal X-ray series, CT scans of the abdomen
and pelvis, CT scan of the chest for the proximal gastric
lesions, complete blood count, renal function tests,
and assessment of performance status (Karnofsky performance score [KPS]). Patients were considered ineligible for entry into the study for the following reasons:
1) physical examination or imaging studies revealing
local or metastatic disease that was judged to preclude
curative gastric resection; 2) hematologic abnormalities including a leukocyte count õ 4000/mm3 and a
platelet count õ 100,000/mm3 , 3) creatinine clearance
õ 60 mL/minute; 4) poor performance status (KPS õ
70); 5) conditions requiring urgent surgery, such as
gastric outlet obstruction or active gastric hemorrhage;
and 6) history of another invasive malignancy, except
basal skin tumors.
Between August 1988 and February 1993, 278
newly diagnosed patients were assessed by the faculty
of the USC Departments of Medicine and Surgery, and
201 were deemed beyond reach of curative resection
on the basis of metastatic or locally invasive disease.
Of the 77 patients eligible for the trial, 9 declined to
participate, 4 were inadvertently not offered entry, and
5 underwent surgery urgently because their presentation suggested an acute abdomen. Data from the remaining 59 patients form the basis for this report.
Preoperative Chemotherapy
Systemic preoperative chemotherapy with cisplatin
and 5-FU commenced on Day 1. Cisplatin at 100 mg/
W: Cancer
Perioperative Gastric Carcinoma Therapy/Crookes et al.
m2 was infused at 1 mg/minute immediately after
which PI 5-FU, 200 mg/m2 /day, was begun and continued for 21 days using an outpatient ambulatory infusion pump attached to an implanted central venous
access device. Intravenous leucovorin, 20/mg/m2 , was
given on the first day of 5-FU therapy and repeated
weekly while the patient was receiving 5-FU. Specific
antiemetic therapy was left to the discretion of individual investigators. On Day 29, the second cycle of cisplatin, PI 5-FU and weekly leucovorin was begun.
Hematologic and clinical toxicity was graded using
the Common Toxicity Criteria. The protocol called for
dose reductions of leucovorin if the patient developed
the 5-FU-related complications of mucositis, diarrhea,
hand-foot syndrome, or bone marrow suppression. Although dose reductions of 5-FU were not a part of
this protocol, if the total granulocyte and/or platelet
counts were õ500/mm3 and 50,000/mm3 , respectively, the second course of 5-FU was omitted, and the
patient underwent surgery after bone marrow recovery.
Cisplatin was eliminated from the second systemic cycle if there were abnormal elevations of serum
creatinine or blood urea nitrogen. Clinically detectable
neurologic toxicity also was cause for cancelling the
second cycle of cisplatin. No dose escalations of the
preoperative agents were allowed.
reconstruction by esophagojejunostomy. When the
gastroesophageal junction was resected, vagal fibers
were interrupted and, therefore, a pyloromyotomy or
pyloroplasty was performed. Some proximal gastric tumors required a thoracoabdominal approach. It was
prospectively decided that if all resected lymph nodes
contained metastatic disease, then the surgery would
not be considered potentially curative.
During the course of this study, data (emanating
primarily from the Japanese Research Society for Gastric Cancer) were published that suggested that extended lymphadenectomy may provide a survival benefit.17 – 19 Consequently, in the latter 21 patients in this
series, the standard dissection of perigastric lymph
nodes (D1 dissection) was extended to include lymph
nodes adjacent to the left gastric artery, common hepatic artery, celic axis, splenic hilum, and splenic artery (D2 dissection).
At the conclusion of the gastric surgery all but
two patients received an intraperitoneal catheter for
postoperative IP chemotherapy. In two patients, this
step was inadvertently omitted, and the catheter was
placed during a separate procedure under local anesthesia prior to beginning IP therapy.
Most patients also received a needle-catheter jejunostomy at the conclusion of the original surgery to
enable postoperative enteral feeding.
IP Chemotherapy
To determine the feasibility of IP chemotherapy after
such extensive gastric surgery, the IP distribution of
fluid instilled through the peritoneal catheter was
studied postoperatively in the first 14 patients by double contrast CT scans using both oral and IP contrast
material. In every case, the study demonstrated adequate distribution of IP contrast, and, therefore, this
test was not performed in subsequent patients.
IP therapy commenced 21 – 42 days after surgery.
A total daily dose of 3000 mg of FUdR was administered IP for 3 consecutive days. The drug was prepared
in 2 L of normal saline containing 4 mEq KCl/l and
warmed to 37 7C. The FUdR solution was allowed to
flow rapidly into the peritoneal cavity by gravity. The
IP solution was drained only if the patient developed
painful abdominal distention. On the fourth day, IP
cisplatin, 200 mg/m2 , was administered in 1 L of normal saline warmed to 37 7C and drained after 4 hours.
Just prior to the IP cisplatin, sodium thiosulfate, 4 mg/
m2 in 250 mL of sterile water, was administered intravenously. After the IP cisplatin, a further dose of sodium thiosulfate, 12 mg/m2 , was given intravenously
over 6 hours. The second cycle of IP chemotherapy
started 3 weeks after the first cycle. Dose reductions
or discontinuation of IP therapy because of toxicity
were permitted and were similar to those outlined in
Gastric resection was scheduled 3 – 4 weeks after completion of preoperative chemotherapy. The response
to chemotherapy was evaluated just before surgery by
repeating the staging investigations previously undertaken to assess eligibility.
The specific surgical procedure performed was determined by the attending surgeon based on pretreatment tumor location. A curative resection was defined
as en bloc removal of the gastric tumor (with macroscopic and microscopic free proximal and distal margins), together with draining lymph nodes and the
attached omentum. In cases of complete response
(CR) to preoperative therapy, the extent of gastric resection was determined by the projected site of the
tumor, based on pretreatment imaging and endoscopy. If, at the same surgery, en bloc resection of the
transverse colon, spleen, distal pancreas, or portion of
the liver was performed, and a clear margin obtained,
the surgery was also considered curative.
In general, tumors not requiring resection of the
gastroesophageal junction (i.e., tumors of the corpus
and antrum) were removed by distal gastrectomy and
reconstruction by gastrojejunostomy. Fundal and cardiac tumors were treated either by proximal gastrectomy with a margin of esophagus and reconstruction
by esophagogastrostomy or by total gastrectomy with
/ 7b55$$1073
04-02-97 14:46:42
W: Cancer
CANCER May 1, 1997 / Volume 79 / Number 9
Protocol Scheme for Therapy
Selected Demographics
Protocol day
Route of
Preoperative day
1 and 29
1 through 22
29 through 50
1, 8, and 15 and 29, 36, and 42
Cisplatin 100 mg/m2
PI 5-FU 200 mg/m2/day
PI 5-FU 200 mg/m2/day
Bolus LV 20 mg/m2
Systemic (i.v.)
Systemic (i.v.)
Systemic (i.v.)
Systemic (i.v.)
Postoperative day
FUdR 3000 mgb
Cisplatin 200 mg/m2
with N sodium thiosulfate
FudR 3000 mgb
Cisplatin 200 mg/m2
with sodium thiosulfate
Systemic (i.v.)
Systemic (i.v.)
i.v.: intravenously; PI: protracted infusion; 5-FU: 5-fluorouracil; LV: leucovorin; FUdR: 5-fluoro-2* deoxyuridine.
Postoperative therapy began 22–43 days after surgery.
Total dose.
the protocol for the preoperative systemic therapy.
The protocol is outlined in Table 1.
Toxicity Evaluation and Response Criteria
Patients were evaluated weekly after entry into the
study. Notation was made of body weight; abdominal,
epigastric, or back pain; general performance status;
complete blood count; and serum creatinine and urea
levels. Response to preoperative chemotherapy was
assessed by serial CT scans to measure focal gastric
wall thickness at the site of the tumor and the size of
lymph nodes draining the gastric tumor; endoscopic
evaluation of the mucosal extent of the tumor; and
histologic determination of residual tumor in the operative specimen. Criteria for a CT partial response (PR),
applicable only in patients with measurable disease
initially, included reduction in gastric wall thickness
at its thickest point by at least 50% and a 50% reduction in the product of the perpendicular greatest dimensions of a single measurable lymph node or the
sum of the products when there were multiple lymph
nodes visualized. A CT CR required normal appearance of the stomach, lymph nodes, and surrounding
tissue planes. An endoscopic response was defined as
complete resolution of all mucosal disease. A pathologic CR mandated total resolution of all cancer
(T0M0N0 ) on histologic examination of the resected
Follow-up Evaluations
After completion of all protocol therapy, patients were
followed jointly by faculty and fellows from the USC
/ 7b55$$1073
04-02-97 14:46:42
Total patients
Age (yrs) (Median, 55 yrs)
Tumor Location
GE junction
Tumor histology
No. of
of total
GE: gastroesophageal.
Surgery Service and the Medical Oncology Division for
symptoms and physical evidence of recurrence. Upper
gastrointestinal endoscopy and CT scans of the chest,
abdomen, and pelvis were not routinely scheduled in
the follow-up regimen. These examinations were performed only if patients developed unexplained weight
loss or symptoms consistent with recurrent gastric carcinoma. A clinical nurse specialist made progress
notes during patient visits to both the medical oncology and the surgical clinics.
Demographic data are tabulated in Table 2. A clearly
defined chief presenting symptom was identified in 50
of the 59 patients (85%). Thirty patients (51%) presented with pain, 10 (17%) with early satiety, and 10
(17%) with chronic nausea and weight loss. The 9 remaining patients (15%) were found to have gastric carcinoma while being evaluated for anemia, acute nausea and emesis, or hematemesis. The median KPS was
Treatment Compliance
All 59 patients entered into the trial received the first
cycle of preoperative systemic chemotherapy, and 58
(98%) received both cycles. Fifty-six patients (95%) un-
W: Cancer
Perioperative Gastric Carcinoma Therapy/Crookes et al.
derwent surgery to remove a gastric adenocarcinoma
within 3 – 4 weeks after the last day of PI 5-FU. Two
patients refused surgery, and one patient died of progressive gastric carcinoma prior to surgery. Of the 40
patients who underwent potentially curative gastric
surgery, 31 received both cycles of IP chemotherapy
with FUdR and cisplatin. Four patients refused the
second course of IP therapy and, therefore, completed
only one cycle. Five patients refused any IP chemotherapy.
Systemic Chemotherapy: Toxicity
Hematologic toxicity
No mortality was associated with the systemic chemotherapy. Despite a schedule of 5-FU and leucovorin that
is generally marrow-protective,20 7 of 59 patients (12%)
developed Grade 4 granulocytopenia after the first cycle
of chemotherapy. Dose reduction or elimination of leucovorin from the second cycle enabled all seven of these
patients to receive the full dose of 5-FU with no more
than Grade 2 granulocytopenia resulting. One patient
exhibited Grade 2 granulocytopenia in connection with
venous catheter-related bacteremia, which developed
after the second course of systemic chemotherapy. The
single patient who failed to complete the 2 cycles of
systemic chemotherapy had a nadir granulocyte count
of 1.2 1 103 /mm3 (Grade 3 toxicity) after the first cycle
without recovery in the alloted period. As per protocol,
the second cycle was cancelled, and the patient underwent surgery 1 week late but without further complications. No patient developed a platelet count õ 78,000/
mm3 . Five patients received transfusion of packed red
blood cells to meet the eligibility criterion of 10 g/dL
of hemoglobin. Eight additional patients received transfusions preoperatively.
Nonhematologic toxicity
The most frequent and troubling nonhematologic
morbidity was cisplatin-induced nausea and emesis.
This problem was especially prevalent in the early part
of the study, affecting 29 of 38 patients (76%), before
odansetron became available. After adding this agent
to the antiemetic regimen, only 7 of the last 21 patients
(33%) had nausea and emesis. One patient developed
mild renal insufficiency (blood urea nitrogen, 29 mg/
dL, and creatinine, 1.9 mg/dL) that, however, did not
complicate subsequent surgery. Stomatitis was observed in 10 patients (17%) during the first cycle of PI
5-FU and leucovorin, but reducing the leucovorin dose
enabled completion of the second cycle of chemotherapy. Seven patients (12%) had mild diarrhea that did
not require any drug dose modifications, and 8 patients (14%) developed mild skin rashes attributed to
the PI 5-FU.
/ 7b55$$1073
04-02-97 14:46:42
Surgery: Mortality and Morbidity
There were two surgical deaths, yielding a 3% surgical
mortality. One patient died of an apparent intraoperative myocardial infarction, and the other died within
10 hours of surgery, most likely from pulmonary embolism (autopsy refused). Postoperative morbidity included three intraabdominal abscesses occurring in
patients whose surgery included splenectomy. Two
patients required reoperation, one for a presumed
anastomotic leak and drainage of an associated abscess and another for wound dehiscence. Two wound
infections healed by secondary intention. One patient
developed postoperative pneumonitis and required
reintubation and ventilatory support. In addition, significant systemic sepsis occurred postoperatively in
three patients, two of whom were infected by Candida
species requiring amphotericin B therapy. In two patients, chemical peritonitis developed as a result of the
patients inadvertently instilling tube feeding preparations into the intraperitoneal catheter instead of into
the feeding jejunostomy catheter.
IP Chemotherapy: Toxicity
Hematologic toxicity
Granulocytopenia was the most significant hematologic toxicity stemming from the IP therapy. The median granulocyte count decreased progressively from
3.6 1 103 /mm3 prior to IP therapy to 1.9 1 103 /mm3
2 weeks after the first IP cycle and to 1.5 1 103 /mm3
2 weeks after the second IP treatment. Four patients
(10%) experienced Grade 4 granulocytopenia (between
0.1 1 103 /mm3 and 0.45 1 103 /mm3) lasting 3 – 5 days.
One of these patients, after recovering from a nadir
granulocyte count of 0.1 1 103 /mm3 , died of peritoneal sepsis. The median pretreatment hemoglobin for
the 35 patients who received the first course of IP
therapy was 11.6 g. The median 3-week posttherapy
hemoglobin was 9.8 g for the 31 patients who received
both cycles of IP chemotherapy. The IP chemotherapy
had almost no clinically significant effect on platelet
production, because only 1 patient developed a platelet count õ 100,000/mm3 . Table 3 outlines the hematologic toxicity for patients who received IP therapy.
Nonhematologic toxicity
Seven patients reported Grade 3 nausea and emesis.
Two patients developed cisplatin-induced Grade 2 peripheral neuropathy at the end of all therapy. Renal
toxicity was not dose-limiting in this trial. The median
serum creatinine for the 31 patients who completed
all planned systemic and IP therapy remained unchanged.
Because of overall toxicity from the IP therapy, 4
of 35 patients refused the second cycle of IP treatment.
W: Cancer
CANCER May 1, 1997 / Volume 79 / Number 9
Hematopoietic Toxicity by Systemic and Intraperitoneal Course
Hemoglobin g/dL
On study (n Å 59)
After i.v. course 1 (n Å 59)
After i.v. course 2 (n Å 58)
After IP course 1 (n Å 35)
After IP course 2 (n Å 31)
Platelet count 1 103/mm3
Total leukocyte count
1 103/mm3
Granulocytes 1 103/mm3
0.1a –5.
i.v.: intravenously; IP: intraperitoneal.
One granulocytopenic septic death.
Therapeutic Response
Thirty-four of the 50 patients (68%) who presented
with a discernible chief presenting symptom reported
partial or complete relief of that symptom after 2 cycles of systemic chemotherapy. Within 2 – 3 weeks after
completing the preoperative regimen, 36 patients had
achieved an increase in body weight of 5% or an increase in serum albumin of 0.5 gm/dL.
Computerized tomography
Fifty-eight patients had CT scans before and after systemic chemotherapy to assess the primary tumor and
draining lymph nodes. Of 41 patients with scans
judged evaluable for response, 17 (41%) had PR by
criteria outlined above. Six of 58 patients (10%) had
disease progression based on CT findings of new
lymph nodes or increased gastric wall thickness. Five
of the six patients judged to have disease progression
could not be resected for cure because of Stage IV
disease. Five patients who met CT criteria for PR had
Stage IV disease at surgery. Figure 1 illustrates a chemotherapy-induced change in the CT appearance of
the primary gastric carcinoma.
Endoscopic evaluation
Forty-seven of 59 patients (80%) underwent a second
endoscopy with biopsy to evaluate response to the
chemotherapy given prior to surgery. Sixteen patients
(27%) had no gross or microscopic tumor visible, and
these findings correlated with pathologic CR in five
patients. In general, patients with less bulky gastric
carcinomas initially tended to have a mucosal CR.
Seven patients with endoscopic responses were not
evaluable by CT examination because their tumors
could not be measured prior to treatment. Four patients were found to have endoscopic evidence of primary tumor progression after chemotherapy.
Overall, 32 of 59 patients (54%) responded to the
/ 7b55$$1073
04-02-97 14:46:42
preoperative chemotherapy as judged by CT scan, endoscopy, or both.
TNM staging
Of the 56 patients taken to surgery, complete staging
information was available for 55 (there was 1 surgical
death). Forty patients (71%) had a resection with curative intent. Five patients (9%) demonstrated pathologic CR to the chemotherapy, with no carcinoma in
the stomach or draining lymph nodes (T0N0M0 ).
Twelve of the 56 patients (21%) were found to be
T1N0M0 or T2N0M0 , or Stage I. Thirteen patients (23%)
were T1N2M0 , T2N1M0 , or T3N2M0 , or Stage II. Ten patients (18%) were T3N1M0 or T3N2M0 , or Stages IIIA
and IIIB, respectively. Fifteen patients (27%) had either
gross metastatic disease or a primary tumor that could
not be completely resected. Table 4 outlines the results
by TNM and stage classification.
Recurrence Data
Nine of 40 patients (23%) with Stage 0-IIIB gastric carcinoma had biopsy-proven recurrences, which were
documented 3 months – 4 years after the last course of
therapy. Of the nine patients whose carcinoma recurred, one presented with ascites, two had pain secondary to retroperitoneal lymph node metastases, two
had ovarian recurrences, two had liver metastases, one
had bilateral pulmonary metastases, and one had rapidly progressing lymphadenopathy in the retroperitoneum, mediastinum, and supraclavicular areas.
All 15 Stage IV patients died of disseminated gastric
carcinoma 4 – 18 months after palliative surgery. The
two patients who refused surgery died of disseminated
disease 32 and 40 months, respectively, after completion of systemic therapy.
Of the 40 patients who underwent potentially curative surgery for Stage O-IIIB disease, 9 (23%) re-
W: Cancer
Perioperative Gastric Carcinoma Therapy/Crookes et al.
FIGURE 1. (a) Computerized tomography (CT) scan of primary gastric tumor prior to chemotherapy. Arrow
points to tumor mass. (b) Follow-up CT scan after chemotherapy, showing reduced thickness of gastric wall
at the site of the tumor (arrow).
curred and 10 (25%) died. Two patients died from
complications of therapy including one postoperative
death and one death from granulocytopenic sepsis
after IP therapy. Eight of the nine patients who recurred died of disseminated disease. One patient who
recurred after 4 years was alive at last follow-up while
responding to the same systemic chemotherapy that
was given preoperatively.
Four of 10 (40%) patients determined to have
/ 7b55$$1073
04-02-97 14:46:42
Stage III disease at the time of surgery died of recurrent
gastric carcinoma (3 Stage IIIB patients and 1 Stage
IIIA patient); 3 of 13 Stage II patients (23%) died, but
only 1 of recurrent gastric carcinoma, and 2 of 12 Stage
I patients (17%) have died of disseminated disease. No
patient with Stage 0 disease recurred.
The median follow-up for all patients enrolled in
this trial was 43 months. Thirty-one of the 59 patients
(53%) were alive at last follow-up. At 2 years, the esti-
W: Cancer
CANCER May 1, 1997 / Volume 79 / Number 9
Kaplan–Meier plot of probability of survival for all gastric
carcinoma patients treated with preoperative systemic chemotherapy followed by postoperative intraperitoneal chemotherapy in University of
Southern California pilot trial.
TNM/Stage Classification (n Å 59)
TNM classification
No. of
Percent of
2 refused OR
1 preop death
1 unstaged a
OR: surgery; preop: preoperative.
Died at surgery; staging incomplete.
mated probability of surviving is 0.64 { 0.06 (based
on the Kaplan – Meier product-limit estimator and
Greenwood’s formula for the standard error), and it is
0.52 { 0.07 at 4 years. The median survival is estimated
as 52 months, with a 95% confidence interval of (25
months, 76/ months) (Figure 2).
Data from modern postoperative adjuvant trials suggest a median survival of approximately 2 years for
patients undergoing successful gastric resection, but
/ 7b55$$1073
04-02-97 14:46:42
when all patients undergoing surgery are considered,
the median survival is usually õ1 year. An earlier trial
of combined pre- and postoperative systemic chemotherapy by Ajani et al.21 in 25 patients with resectable
gastric carcinoma yielded a 15-month median survival, but 13 patients (52%) developed peritoneal carcinomatosis. Atiq et al.13 treated patients after resection
of Stage IIIA and Stage IIIB tumors with systemic chemotherapy followed by IP 5-FU. Peritoneal metastases
occurred less often than in historic controls but still
with disturbing frequency.
These discouraging data led the authors to depart
from the methodology of usual clinical trials and design a pilot chemotherapy trial with preoperative and
postoperative components despite the absence of definitive evidence that either decreases recurrence rates
or improves survival. The authors consider the postoperative locoregional regimen employed here to be a
key component of their program, because prior chemotherapy trials have not had a substantial impact on
peritoneal recurrences.
Determining the efficacy of preoperative chemotherapy is difficult. Although the majority of the patients in the current study reported subjective symptomatic improvement, verification of primary tumor
response by objective tests such as endoscopic visualization, CT imaging, or ultrasonography is not always
possible because reproducible measurements are often beyond the limits of these techniques.22,23 Nevertheless, the authors judged 41% of the current study
patients to have had a CT-demonstrated response and
27% an endoscopic-determined response, yielding an
overall response rate of 54%, including 5 complete
pathologic responses. After preoperative chemotherapy, 51% of the patients in the current study were
classified at the time of surgery to be Stage 0-II. This
is in striking contrast to an only 20% incidence of Stage
I and II disease observed in a similar population of
patients treated in the 5 years preceding the current
trial at the Los Angeles County-USC Medical Center.1
This apparent downstaging among patients undergoing surgery at this institution may be attributable to
the preoperative chemotherapy. In this series, only
three of nine recurrences initially developed within the
peritoneal cavity. This low frequency of locoregional
recurrences may be due to the IP therapy and the use
of the multiple, noncross-resistant agents employed
in the study protocol.
The most striking feature of this analysis is the
excellent median survival observed (ú4 years). The
results of this current study with a larger number of
patients followed for a longer period confirm those in
the authors’ preliminary report.16 The authors attribute this apparently improved survival to the chemotherapy program rather than to the extended lymph-
W: Cancer
Perioperative Gastric Carcinoma Therapy/Crookes et al.
adenectomy employed in the last third of the patients
accrued because the improved median survival was
noted in the earlier cohort of patients not undergoing
D2 dissections.16
The authors accept that a single-arm pilot trial
cannot definitively determine the impact of chemotherapy on survival from gastric carcinoma, but the
encouraging salutary effects observed here, using a
drug regimen that can be safely administered to the
average patient with gastric carcinoma, indicate that
this aggressive program should be put to the test of a
prospectively randomized trial.
Leichman L. 1987 Cancer Program Report: gastric cancer, a
survey of two years for the Los Angeles County / University
of Southern California Medical Center. Annual Report
Diehl JT, Hermann RE, Cooperman AM, Hoerr SO. Gastric
carcinoma: a ten year review. Ann Surg 1983;198:9–12.
Coombes RC, Schein PS, Chilvers CED, Wil J, Beretta G, Bliss
JM, et al. A randomized trial comparing adjuvant fluorouracil, doxorubicin and mitomycin-c with no treatment in operable gastric cancer. J Clin Oncol 1990;8:1362–9.
Macdonald JS, Fleming TR, Peterson RF, Berenberg JL, McClure S, Chapman RA, et al. Adjuvant chemotherapy with
5FU, Adriamycin and Mitomycin-c (FAM) versus surgery
alone for patients with locally advanced gastric adenocarcinoma: a Southwest Oncology Group Study. Ann Surg Oncol
Wilke H, Preusser P, Fink U, Gunzer U, Meyer J, Meyer JR,
et al. Preoperative chemotherapy in locally advanced and
nonresectable gastric cancer: a phase II study with etoposide, doxorubicin and cisplatin. J Clin Oncol 1989;7:1318–
Leichman L, Herskovic A, Leichman CG, Lattin PB, Steiger
Z, Tapazaglou E, et al. Nonoperative therapy for squamous
cell cancer of the esophagus. J Clin Oncol 1987;5:365–70.
Wisbeck WM, Becher EM, Russell AH. Adenocarcinoma of
the stomach: autopsy observations with therapeutic implications for the radiation oncologist. Radiother Oncol
Gunderson LL, Sosin H. Adenocarcinoma of the stomach:
areas of failure in a reoperation series (second or symptomatic look): clinicopathologic correlation and implications for
adjuvant therapy. Int J Radiat Oncol Biol Phys 1982;8:1–11.
Leichman CG, Leichman L, Spears CP, Rosen PJ, Muggia F,
Jeffers S, et al. Biological modification of protracted infusion
of 5-fluorouracil with weekly leucovorin. A dose seeking clinical trial for patients with disseminated gastrointestinal cancers. Cancer Chemother Pharmacol 1990;26:57–61.
/ 7b55$$1073
04-02-97 14:46:42
10. Leichman L, Steiger Z, Seydel HG, Dindogru A, Kinzie J,
Toben S, et al. Preoperative chemotherapy and radiation
therapy for patients with cancer of the esophagus: a potentially curative approach. J Clin Oncol 1984;2:75–9.
11. Leichman L, Berry B. Cisplatin therapy for adenocarcinoma
of the stomach. Semin Oncol 1991;18(Suppl 3):25–33.
12. Muggia FM, Chan KK, Russell C, Colombo N, Speyer JL,
Sehgal K, et al. Phase I and pharmacologic evaluation of
intraperitoneal 5-fluoro-2*-deoxyuridine. Cancer Chemother
Pharmacol 1991;28:241–50.
13. Atiq OT, Kelsen DP, Shiu MH, Saltz L, Tong W, Niedzwiecki
D, et al. Phase II trial of postoperative adjuvant intraperitoneal cisplatin and fluorouracil and systemic fluorouracil
chemotherapy in patients with resected gastric cancer. J Clin
Oncol 1993;11:425–33.
14. Howell SB, Pfeifle CE, Wung WE, Olshen RA, Lucas WE, Yon
JL, et al. Intraperitoneal cisplatin with systemic thiosulfate
protection. Ann Intern Med 1982;97:845–91.
15. Markman M, Reichman B, Hakes T, Jones W, Lewis JL, Rubin
S, et al. Responses to second-line cisplatin-based intraperitoneal therapy in ovarian cancer: influence of a prior response to intravenous cisplatin. J Clin Oncol 1991;9:1801–
16. Leichman L, Silberman H, Leichman CG, Spears M, Ray M,
Muggia FM, et al. Preoperative systemic chemotherapy followed by adjuvant postoperative intraperitoneal therapy for
gastric cancer: a University of Southern California Pilot Program. J Clin Oncol 1992;10:1933–42.
17. Hagen JA, Peters JH, DeMeester TR. Superiority of extended
en bloc esophagogastrectomy for carcinoma of the lower
esophagus and cardia. J Thorac Cardiovasc Surg 1993;
18. Japanese Research Society for Gastric Cancer. The general
rules for the gastric cancer study in surgery and pathology.
Jpn J Surg 1981;11:127–37.
19. Smith JW, Shiu MH, Kelsey L, Brennan MF. Morbidity of
radical lymphadenectomy in the curative resection of gastric
carcinoma. Arch Surg 1991;126:1469–73.
20. Leichman CG, Leichman L, Spears CP, Rosen PJ, Jeffers S,
Groshen S, et al. Prolonged continuous infusion of fluorouracil with weekly bolus leucovorin: a phase II study in patients with colorectal cancer. J Natl Cancer Inst 1993;85:41–
21. Ajani JA, Ota DM, Jessup M, Ames FC, McBride C, Boddie
A, et al. Resectable gastric carcinoma: an evaluation of preoperative and postoperative chemotherapy. Cancer 1991;
22. Lightdale CJ, Botet JF, Brennan MF, Shiu MH, Coit DG. Endoscopic ultrasonography (EUS) compared to computerized
tomography (CT) for pre-operative staging of gastric cancer
[abstract]. Gastrointest Endosc 1989;35:154–5.
23. Tio TL, Den Hartog Jager FC, Tytgat GNJ. The role of endoscopic ultrasonography in assessing local resectability of oesophagogastric malignancies. Accuracy, pitfalls, and predictability. Scand J Gastrenterol 1986;123(Suppl):78–86.
W: Cancer
Без категории
Размер файла
192 Кб
Пожаловаться на содержимое документа