Cytoreductive Surgery and Intraperitoneal

7/31/2019 Cytoreductive Surgery and Intraperitoneal

1/11

Cytoreductive Surgery and IntraperitonealHyperthermic Chemotherapy for Peritoneal SurfaceMalignancy: Experience with 501 Procedures

Edward A Levine, MD, FACS, John H Stewart IV, MD, Gregory B Russell, MS, Kim R Geisinger, MD,Brian L Loggie, MD, FACS, Perry Shen, MD, FACS

BACKGROUND: Peritoneal dissemination of abdominal malignancy (PSD) has a clinical course marked by bowelobstruction and death. We have been using aggressive cytoreductive surgery with intraperito-neal hyperthermic chemotherapy (IPHC)to treat PSD. Thepurpose of this article was to reviewour experience with IPHC.

STUDY DESIGN: A prospective database of patients undergoing IPHC has been maintained since 1991. Patientswere uniformly evaluated and treated. Demographics, performance status, resection status,primary site, and experience quartile were compared with outcomes. Univariate and multivar-

iate analyses were performed.RESULTS: A total of 460 patients underwent 501 IPHC procedures. Average age was 53.0 years, and50.4% were women.The 30-day mortality rate was 4.8%, the complication rate was 43%, andmedian hospital stay was 9 days. Median followup was 55.4 months, median survival was22.2 months, and 5-year survival rate was 27.8%. Factors correlating with improved survival

were performance status (p 0.0001), primary tumor (p 0.0001), resection status(p 0.0001), complications (p 0.002), previous IPHC (p 0.006), and experience quar-tile (p 0.031). On multivariate analysis, primary tumor site, performance status, resectionstatus, and development of complications (p 0.001) predicted outcomes.

CONCLUSIONS: Our experience demonstrated that preoperative criteria for better outcomes include primarytumor site and performance status. Completeness of resection and development of postopera-tive complications are also crucial, and outcomes have improved over time. Cytoreductive

surgery and IPHC represent substantial improvements in outcomes compared with historicseries and best-available systemic therapy. Longterm survival is possible for selected patientswho undergo the procedure. ( J Am Coll Surg 2007;204:943955. 2007 by the AmericanCollege of Surgeons)

Disseminated malignant peritoneal surface disease (PSD),or carcinomatosis, has traditionally and justifiably beenapproached with therapeutic nihilism. Patients typicallyprogress to death from bowel obstruction in less than a

year.1 PSD results from intracavitary dissemination of tu-mor from a variety of primary pathologic lesions. Suchfindings are all too common for gastrointestinal and ovar-ian carcinomas, and are also seen with unusual malignan-cies such as sarcoma, mesothelioma, and urachal carci-noma. Frequently, PSD is confined to the peritoneal cavitywithout extraabdominal disease. So, a regional approach toselected patients with PSD is reasonable.

In the 1980s, aggressive treatments of peritoneal surfacemalignancies were explored in an attempt to benefitthrough multimodality techniques. Centers explored treat-ment options such as peritonectomy procedures,2 IP injec-tion of anticancer drug OK432,3 intracavitary immuno-therapy, photodynamic therapy,4,5 IP hyperthermicchemotherapy (IPHC), and early postoperative IPchemotherapy.6-8 Over the past decade, there has been everincreasing interest in such regional therapy for PSD. This

Competing Interests Declared: None.

A portion of this work was supported by the Robert Welborne fund.Presented at the Southern Surgical Association 118th Annual Meeting, WestPalm Beach, FL, December 2006.

Received December 2, 2006; Accepted December 15, 2006.From the Surgical Oncology Service, Department of General Surgery (Le-vine, Stewart, Shen), the Section on Biostatistics, Department of PublicHealth Sciences (Russell), and the Department of Pathology (Geisinger),Wake Forest University School of Medicine, Winston-Salem, NC; and theSurgical Oncology Section, Department of General Surgery, Creighton Uni-versity School of Medicine, Omaha, NE (Loggie).Correspondence address: Edward A Levine, MD, Department of GeneralSurgery, Surgical Oncology Service, Wake Forest University School of Med-icine, Medical Center Blvd, Winston-Salem, NC 27157.

943 2007 by the American College of Surgeons ISSN 1072-7515/07/$32.00

Published by Elsevier Inc. doi:10.1016/j.jamcollsurg.2006.12.048


2/11

has been additionally stimulated by publication of a pro-spective randomized trial for PSD from colorectal sources9

and recent successes with IP chemotherapy for PSD fromovarian cancer.10,11

The optimal management of patients with PSD is a mat-ter of intense debate. Systemic chemotherapy for PSD islimited, in part, to its limited entry into the peritoneum.

Localization of tumor within the peritoneum without dis-tant metastasis makes an aggressive regional approach at-tractive. Several groups have treated peritoneal surface ex-tension of appendiceal tumors with debulkingprocedures.12-14 But it is clear that these procedures arefrequently unable to remove all microscopic tumors.

Our approach to selected patients with PSD has been tocombine aggressive cytoreductive surgery (CS), with thegoal of resecting all gross disease, with chemoperfusion toaddress microscopic residual tumor. A chemotherapy per-fusion done at the same time as CS has several advantages.First, intracavitary chemotherapy achieves drug levels far

greater than can be obtained with even the most aggressivesystemic administration, which may overcome relative in-trinsic drug resistance. Next, after CS, all peritoneal sur-faces are exposed, allowing for better drug distribution(versus postoperative) as all adhesions are lysed. Addition-ally, the single intraoperative dose eliminates major com-pliance and tolerance issues encountered with postopera-tive administration of several cycles of treatment.10,11,15Therationale for hyperthermia is based on laboratory studiesshowing synergy with certain drugs.

We have previously reported our early experience8 andsubsets of patients treated with CS and IPHC for PSD

from appendiceal,16,17

colorectal,18,19

gastric,20

and smallbowel21 carcinomas and mesothelioma.22 This article re-views our prospective database of patients undergoing CSand IPHC for PSD to evaluate our experiences with thefirst 501 procedures.

METHODS

Patients who underwent CS and IPHC for PSD at WakeForest University School of Medicine Baptist Hospital be-tween 1991 and 2006 were identified from a prospective

database. This database and analysis has been approved bythe Institutional Review Board at Wake Forest University.All patients were evaluated in the surgical oncology clinicspreoperatively. Evaluations included, at a minimum, acomplete history, examination, pathologic review, CT im-aging, blood counts, and renal and liver functions. To beconsidered for CS and IPHC, patients needed to have nor-mal organ function (serum creatinine 2 mg/dL, alkalinephosphatase and serum aspartate transaminase or alaninetransaminase less than 3 times the upper limit of normal,white blood cell count 4,000/mm3, and plateletcount 100,000 mm3. Evaluation of preoperative CTimaging focused on the absence of extraabdominal metas-tasis, parenchymal hepatic metastasis (hepatic surface le-sions allowed), bulky small bowel disease, multistationbowel obstruction, and ureteral or biliary obstruction. Tu-mors were categorized according to the primary site of or-igin. Before CS and IPHC, patients had their pathologyreviewed by the Wake Forest University Department ofPathology. This was compared with final pathology fromspecimens garnered at the time of CS to reach a final diag-nosis for the database. Patients with bulky pelvic disease ormultiple previous pelvic procedures were routinely consid-ered for urologic consultation for ureteral stent placementat the start of the procedure to facilitate retroperitonealdissection. Clinical data on all patients were recorded in adatabase and maintained by a dedicated data managementunit.

Cytoreductive surgery

The goal of cytoreductive surgery was removal of all grossdisease in all patients. CS consisted of the removal of grosstumor and involved organs, peritoneum, or tissue deemedtechnically feasible and safe for the patient. This includedroutine supracolic omentectomy in all patients when it hadnot been previously performed. Peritonectomy procedureswere performed as indicated.2Any tumors adherent or in-vasive to vital structures that could not be removed werecytoreduced using the cavitational ultrasonic surgical aspi-rator (CUSA, Valleylab). The resection status of patientswas judged after CS using the following classification: R0,complete removal of all visible tumor and negative cyto-logic findings or microscopic margins; R1, complete re-moval of all visible tumor and positive postperfusion cyto-logic findings or microscopic margins; R2a, minimalresidual tumor, nodule(s) measuring 0.5 cm; R2b, grossresidual tumor, nodule 0.5 cm but 2 cm; and R2c,extensive disease remaining, nodules 2 cm.

Intraperitoneal hyperthermic chemotherapy

Near the completion of CS, patients were passively cooledto a core temperature of approximately 34 C to 35 C by

Abbreviations and Acronyms

CS cytoreductive surgeryGIST gastrointestinal stromal tumorIPHC intraperitoneal hyperthermic chemotherapyMMC mitomycin C

OS overall survivalPMP pseudomyxoma peritoneiPSD peritoneal surface malignant disease

944 Levine et al Treatment of Peritoneal Surface Malignancy J Am Coll Surg


3/11

passive measure (ie, not warming airway gases or IV solu-tions and cooling the room). After CS was completed, peri-toneal perfusion inflow and outflow catheters were placedpercutaneously into the abdominal cavity. Temperatureprobes were placed on the inflow and outflow catheters justoutside the exit sites from the abdomen. The abdominalskin incision was closed temporarily with a running cuta-neous suture to prevent leakage of peritoneal perfusate. Aperfusion circuit was established with approximately 3 L ofRingers lactate. Flow rates of approximately 600 to 1,000mL/min were maintained using a roller pump managed bythe perfusionist. The circuit continued through a singleroller pump, through a heat exchanger, and then to thepatient.

Constant temperature monitoring was performed at alltemperature probes. Once a stable perfusion circuit wasestablished and outflow temperature exceeded 38.5 C, thechemotherapy was introduced into the perfusion circuit. Amaximum inflow temperature of 42.5 C was toleratedduring perfusion, with a target outflow temperature at thepelvis of 40 C. The abdomen was gently massagedthroughout perfusion to improve drug distribution to allperitoneal surfaces. Total planned perfusion time after theinitial addition of mitomycin C (MMC) was 120 minutes.Although several chemotherapeutic agents were used, mostpatients received MMC. The MMC was dosed based onvolume of perfusate necessary to establish a stable circuit(typically 3 L). When MMC was used, 30 mg was added tothe perfusate at initiation of the IPHC; at 60 minutes, anadditional 10 mg of MMC was added to keep MMC per-fusate concentrations higher than 5 g/mL. In certain pa-tients (elderly individuals, those with extensive previouschemotherapy, those with inanition or poor performancestatus, and patients having extensive peritoneal strippingduring oepration), reductions in the dose of MMC (to 30mg total) or perfusion time (to 60 to 90 minutes) weremade because of concerns about potential toxic effects.Other chemotherapeutic agents were also used based onprimary tumor site and previous systemic therapy. Since2004, we have used cisplatinum 250 mg/m2 with sodiumthiosulfate for mesothelioma patients and cisplatinum or

carboplatinum for ovarian cancer patients. Sarcoma pa-tients (and gastrointestinal stromal tumor [GIST] patientsbefore the introduction of imatinib) were perfused withMMC with or without mitoxantrone.

Clinical followup

Clinical followup occurred at 1 month, and at least every6 months thereafter for up to 5 years. After 5 years from thelast IPHC, followup was annual. Blood counts, liver func-tions and tumor markers (as appropriate), and abdominaland pelvic CT scans with oral and IV contrast, were ob-

tained with each followup visit and when clinically indi-cated. Patients were typically followed jointly with medicaloncologists. Some patients received systemic chemother-apy at the discretion of their medical oncologists.

Statistical analysis

All data were collected prospectively; descriptive statisticswere generated for all measures, including means, ranges,and standard deviations for continuous measures and fre-quencies and proportions for categorical data. Overall sur-

vival (OS) was calculated from the date of CS and IPHC tothe last known date of followup or date of death. Estimatesof survival were calculated using the Kaplan-Meier(product-limit) method; analysis using Cox proportionalhazards was performed on all pertinent clinicopathologicvariables to determine each ones association with survival.Group comparisons of OS were performed using the ap-proximate chi-square statistic for the log-rank test. Addi-tionally, the Cox proportional hazards regression modelwas used in a stepwise fashion to perform a multivariateanalysis of clinicopathologic factors to determine an overallmodel of independent predictors of OS. Statistical signifi-

cance was defined as a p value

0.05.

RESULTS

Patients and clinicopathologic features

A total of 501 IPHC procedures were performed on 460patients with peritoneal surface malignancy who under-went the procedures between December 30, 1991 and June8, 2006. This study was approved by our Institutional Re-view Board. Patient demographics and baseline data arelisted in Tables 1 and 2. The mean age was 53.0 12.7

Table 1. Clinicopathologic Data for 460 Patients Undergo-ing 501 Cytoreductive Surgeries and Intraperitoneal Hyper-thermic Chemotherapy for Peritoneal Surface Disease

Characteristic n %

Gender

Male 232 49.6

Female 228 50.4Race

Caucasian 406 90.2

African-American 39 8.7

Other 15 1.1

Performance status (ECOG)

0 160 34.8

1 213 46.3

2 67 14.6

3 18 3.9

4 2 0.4

ECOG, Eastern Cooperative Oncology Group.

945Vol. 204, No. 5, May 2007 Levine et al Treatment of Peritoneal Surface Malignancy


4/11

years (range 15 to 87 years of age), with a slight femalepreponderance. The median intensive care unit and hospi-tal stays were 2 and 9 days, respectively. Second IPHC wasperformed on 41 selected patients for recurrent or persis-tent disease,23 with 4 patients undergoing 3 procedures.20.1% of patients had an ileostomy (11.7%) or colostomy(8.4%) created as part of the CS. The organs resected aspart of the CS are listed in Table 3. The median hospitalstay was 9 days, with an average of 15.3 (17.9) days. Mostpatients were admitted to the ICU, with an average stay of2 days.

Primary sites of origin for the patients were as follows:adrenal, 1 (0.2%); appendix, 163 (35.4%); colorectal, 133(28.9%); gall bladder, 3 (0.7%); gastric, 42 (9.1%); gastro-intestinal stromal tumor (GIST), 9 (2%); liver, 1 (0.2%);mesothelioma, 27 (5.9%); ovary, 46 (10%); pancreas (cys-tic neoplasm and intrapancreatic mucinous tumor), 5(1.1%); sarcoma, 11 (2.4%); small bowel, 6 (1.3%); ura-chal, 5 (1.1%); and unknown, 8 (1.7%). The median sur-vival (months) was considerably different by site of originas follows: appendix, 63.5; colorectal, 16.4; gastric, 6.1;mesothelioma, 27.1; ovary, 28.5; and sarcoma, 28.1(p 0.0001). For other histologic sites of origin, the serieshad too few patients for meaningful analysis.

Operative and perfusion data are summarized in Table 2.The organs resected are shown inTable 3. The length of theoperation (range 250 to 1,080 minutes) was dependent onthe extent and location of disease at exploration, but itaveraged just under 10 hours. The quantity of residualdisease was recorded by the primary surgeon and wasscored according to the R status for residual disease.24 Theresection status of all patients undergoing IPHC is listed in

Table 2. Resection status was a marked predictor of survival(p 0.0001). For the purposes of survival calculations, R0and R1 were combined because of the difficulties clearlyseparating them in the setting of PSD.

Mobidity and mortality

The 30-day postoperative morbidity and mortality rateswere 43.1% and 43.9%, respectively. Twenty-two patientsin this study died within 30 days of IPHC. Wound infec-tion, hematologic toxicity, sepsis, respiratory failure, anas-tomotic leak, pneumonia, and enterocutaneous fistula ac-counted for the majority of the postoperative

complications in this cohort of patients. Patients who ex-perienced a complication had poorer survival than thosewho did not (p 0.002). This difference remained pro-nounced on multivariate analysis. Complications were lesscommon in patients undergoing R0/1 resections whencompared with patients with more residual disease(p 0.044).

Experience over time

To evaluate our experience over time, we divided our pa-tient experience into 4 time periods (quartiles) of 115 pa-

Table 2. Operative and Perfusion Data for Patients Under-going Cytoreductive Surgery and Intraperitoneal Hyperther-mic Chemotherapy for Peritoneal Surface Disease

Characteristic n Value

Length of operation, min

Mean 560 175

Range 2501,080Resection status

R0-1 294

R2a 115

R2b 65

R2c 84

Length of hyperthermicchemoperfusion, min

Mean 111.4 19.4

Range 60120

Median 120

Number of IPHC procedures

1 4602 37

3 4

Outflow temperature ofperfusate, C

Mean 40.2 .8

Range 3944

Median 40

IPHC, intraperitoneal hyperthermic chemotherapy.

Table 3. Organs Resected as Part of the Cytoreductive Sur-gery in Addition to Peritoneal Resections

Resected organ n %

Diaphragm 28 5.6

Colon 215 42.9

Rectum 39 7.8

Small bowel 176 35.1Stomach 60 12.0

Spleen 166 33.1

Uterus 29 5.8

Ovaries 73 14.6

Gallbladder 89 17.8

Pancreas 37 7.4

Appendix 45 9.0

Omentum 296 59.1

Kidney 9 1.8

Lung 1 0.2

Liver 33 6.6

Bladder 5 1.00

Omentectomy (supracolic) was performed routinely if not previously re-sected.



5/11

tients each. The median survival rates in months for thefour quartiles were 16.2%, 18.2%, 30.5%, and 30.0%,respectively, p 0.031 (Fig. 5). But the median length ofstay was precisely 9 days in each of the quartiles. The mor-tality rate ranged from 2.6% to 7.0% over thefour quartileswithout marked differences.

Survival and followup

For the cohort of 460 patients with a median followup of55.4 months, the median OS was 22.2 months. One, 3-,and 5-year OS rates were 66.8%, 40.0%, and 27.8%, re-

spectively (Fig. 1). The survival rates included operativemortality. A univariate analysis of clinicopathologic factorswas performed to identify singularly pronounced prognos-tic factors associated with OS after CS and IPHC for PSD.Multivariate analysis of factors affecting survival was per-formed through a stepwise regression technique.This anal-ysis allowed for all variables regardless of level of impor-tance in the univariate analysis. The Cox proportionalhazards regression model found that four clinicopathologicfactors were independent predictors of OS: tumor histol-ogy, resection status, complications, and performance sta-tus (Table 4). Figures 1 to 5 depict the Kaplan-Meier actu-

arial survival curves for these factors.

DISCUSSION

CS and IPHC represent a substantial operative undertak-ing for both patient and surgeon. Average operative timesare approximately 10 hours, with long ICU and hospitalstays that consume substantial resources. Morbidity andmortality are severe, and straightforward preoperative dis-cussions with the patient and family are mandatory. Butproperly selected patients have a real chance at longterm

survival rarely, if ever, realized without such aggressive ef-forts. We have avoided addressing PSD from hepatic biliaryand pancreatic sources principally because of the difficultyin obtaining control of the primary lesion. We currentlyconsider patients with gastric cancer only if R0/1 resectioncan be anticipated.

Systemic chemotherapy for peritoneal surface malignan-cies is limited in effect by its limited entry into the perito-neum. Any systemic chemotherapy for IP disease mustovercome the plasma-peritoneal partition. Pharmacoki-netic studies have confirmed the presence of thisperitoneal-plasma partition by demonstrating that drugsdelivered into the peritoneal cavity have a clearance that isinversely proportional to the square root of its molecularweight.25-27 The addition of IP chemotherapy can beviewed as a tool to overcome this drug resistance and tox-icity associated with systemic administration. Because ofthis partition, drugs without lipophilic properties and highmolecular weights have optimal characteristics for IP ap-plication. The pharmacokinetic advantage of IP perfusioncan be seen by the area-under-the-curve ratios of peritonealfluid to plasma that favor retention of drug in theperitoneum.28-34

In addition to the pharmacokinetic advantage that IPchemotherapy infusion after maximal tumor debulking of-fers, the addition of hyperthermia affects cell membranes,cytoskeletons, synthesis of macromolecules, and DNA re-pair mechanisms.35,36 Our institution (and others) has used

Figure 1. Overall survival for patients treated with cytoreductive

surgery and intraperitoneal hyperthermic chemotherapy.

Table 4. Univariate and Multivariate Analyses of Clinicopath-ologic Variables, Based on Stepwise Regression Analysis

Variables p Value

Univariate

Race 0.87

Dose of mitomycin C 0.82

Resection status 0.0001Gender 0.24

Complications 0.002

Length of operation 0.12

Previous CS and IPHC 0.006

Age 0.19

Temperature of perfusate 0.80

Length of chemotherapy 0.22

Primary tumor histology (site) 0.0007

Performance status 0.0001

Multivariate

Resection status 0.001

Performance status 0.0012Primary tumor histology (site) 0.0001

Complications 0.0006

CS, cytoreductive surgery; IPHC, intraperitoneal hyperthermic chemother-apy.



6/11

MMC primarily. The synergy between MMC and hyper-thermia occurs independent of the cell cycle, allowing formajor tumoricidal activity with brief exposures.37-39 In thisstudy, we evaluated various clinical, treatment, and patho-logic characteristics that potentially have an impact on sur-vival for patients undergoing CS and IPHC with MMC forperitoneal surface neoplasms.

There is a paucity of data about the utility of systemictherapy for PSD in general and for appendiceal tumors,particularly the more common low grade tumors. So, the

foundation of treatment for PSD of appendiceal malignan-cies remains aggressive surgical cytoreduction followed by

hyperthermic peritoneal perfusion. Removal of bulk dis-ease is imperative because even the most ambitious perfu-sion strategies penetrate only 5 mm into peritoneal sur-faces. So aggressive cytoreduction allows hyperthermicchemoperfusion to treat the microscopic or small volumeresidual.

Pathologic characteristics clearly have an impact on theclinical outcomes of patients with PSD. For appendicealtumors, surgical series, patients with low grade mucinous

carcinoma peritonei (also described as disseminated perito-nealadenomucinosis or pseudomyxoma peritonei [PMP]) ex-perience better clinical outcomes than do those with higher

Figure 2. Overall survival by primary tumor site. Differences are

significant: p 0.0001. Figure 3. Overall survival by preoperative performance status. Dif-ferences are significant: p 0.0001.

Figure 4. Overall survival by resection status, p 0.0001.

Figure 5. Overall survival by quartile of experience. Differences are

significant: p 0.031.



7/11

grade nonmucinous appendiceal malignancies.40-46 In thisstudy, the survival rate in patients with high-grade lesions wassubstantially lower than for those with low-grade PSD. This isnot an unexpected finding based on the biologic and molec-ular differences between low- and high-grade nonmucinous

appendiceal tumors.

16,17

PMP does not typically metastasizebeyond the abdominal or pelvic cavity. A recent correlativestudy found that appendiceal tumors associated with PMPhave lower proliferative and apoptotic indices than colorectalepithelial neoplasms do. In addition, PMP-producing tumorshavedecreasedcelladhesionmoleculeexpression.42Such stud-ies are important in determining the biologic underpinningsof PMP.

PMP has been considered the classic indication forIPHC. Five-year survivals after IPHC for PMP haveranged between 66% and 97%, and our experience isconsistent with those results.43-46 Tumor histology is an

independent prognostic factor in the subset of patientswith PMP. The outcomes with the low-grade dissemi-nated peritoneal adenomucinosis and intermediate/hybrid histology were considerably better than those ofhigh-grade peritoneal mucinous carcinomatosis in theoriginal description of the histologic subtypes of PMP.47

Later, data from the Washington Hospital Center sug-gested that patients with low-grade disseminated perito-neal adenomucinosis had a better prognosis than thosewith mucinous adenocarcinoma or intermediate histol-ogy after exploratory laparotomy and cytoreduction.43Asubsequent study revealed a difference in 5-year survival

between patients with low-grade (64%) and those withhybrid/high-grade (54%) treated with CS and IPHC(p 0.05). There were, however, very few peritonealmucinous carcinomas (3 of 36) in the followup study.45

A criticism of past work evaluating CS and IPHC forPNP is that only benign or low-grade lesions were in-cluded in these studies, so confounding factors such astumor biology and patient selection were not accountedfor in these earlier studies. We believe that the behaviorof PMP/appendiceal carcinoma is best described simplyas low- and high-grade rather than using a more cum-bersome three-tier classification.16,17 This clearly dem-

onstrates the differences in tumor biology among thesehistologic subgroups of PMP.

We reported in a earlier study of patients with peritonealcarcinomatosis (from a variety of histologies) that patientswith Eastern Cooperative Oncology Group (ECOG) per-formance scores of 2 to 3 had considerably poorer overallsurvival (median survival of 9.5 months) than patients withperformance scores of 0 or 1 (median survival of21.7 months, p 0.02).18 This study supports our previ-ous findings in a wide variety of tumor types. It also high-

lights the importance of evaluating candidates for the pro-cedure while they are medically fit to undergo such a large-scale intervention. In this study, preoperative performancestatus notably affected survival after CS and IPHC. Pa-tients with performance scores of 0 had median survival

four times that of Eastern Cooperative Oncology Group 2patients; those with poorer performance status had mediansurvival of 6 months (p 0.001). So we select patientsfor IPHC with Eastern Cooperative Oncology Groupscores of 2 or better. Despite recent improvements in sys-temic therapy for colorectal cancer, treating patients withsecond line therapy while their performance status declinesmay deprive candidates of the opportunity to be salvagedwith CS and IPHC.

Patients undergoing complete CS before IPHC had su-perior outcomes compared with those who underwent in-complete CS, regardless of the primary lesion site. The

3-year survival rate for patients with a R0/1 resection was59.9 3.9%; 3-year survivals for patients with R2a, R2b,and R2c resections were 38.7 5.3%, 15.0 5.3%,14.0 4.2%, respectively (p 0.0001). This findingconfirms data from our institution, and others, that dem-onstrate an important survival advantage for patients un-dergoing R0/R1 resection compared with those having R2resections.18,48,49 In a review of 506 patients, Glehen andcoworkers50 analyzed the survival of patients with perito-neal surface malignancies from colorectal primary tumorsundergoing incomplete CS followed by IPHC, and foundthat this treatment paradigm resulted in limited longterm

survival. Patients who are unable to undergo major CS(R2a or better) at laparotomy may be spared the potentialtoxicity of IPHC.

Surgical resection remains the primary mode of therapyfor colon and rectal cancer. Treatment options for patientswith unresectable metastatic disease have improved consid-erably in the past few years. Patients with stage IV colorec-tal cancer treated with newer combinations of cytotoxicchemotherapy,51 biologic agents, or both52 have resulted inan unprecedented median survival of approximately20 months. But such therapeutic combinations are not anoptimal treatment strategy for all categories of stage IV

disease. We are unaware of patient survival beyond 5 yearswith medical therapy of any kind alone. Patients undergo-ing CS and IPHC had a 5-year survival of 17%; thoseundergoing R0/1 resections had survivals twice that.19 Thisis consistent with results from other high volumecenters.50,53

This experience is supported by a randomized trial fromthe Netherlands comparing palliative surgery with chemo-therapy to CS and IPHC with the same systemic chemo-therapy.9 That randomized trial found a doubling of sur-



8/11

vival for patients treated with CS and IPHC.9 So we concurwith the consensus statement in that systemic therapyalone is no longer appropriate for patients with limitedperitoneal dissemination from a primary or recurrent coloncancer.54 The surgical management of peritoneal surface

malignancies of colorectal origin with CS and IPHC hasbeen clearly defined and continues to improve.This aggres-sive strategy has resulted in longterm, disease-free survivalrates that are unprecedented in the literature. Although atthe cost of severe morbidity, properly selected patients havea real opportunity for survival in a situation previouslyapproached with purely palliative intent.

Primary peritoneal mesothelioma is a much less com-mon entity than pleural malignancy.48 Although the mo-lecular characteristics of peritoneal disease differ onlyslightly from the pleural disease, the clinical courses aredisparate.55,56 Peritoneal disease typically presents with as-

cites, abdominal pain, and eventually, bowel obstruction.The disease tends to remain within the abdominal cavityuntil late in the course, and distant metastasis is distinctlyuncommon, making this an excellent candidate for CS andIPHC. We and others have previously reported our expe-rience with this modality,22,57-59 which represents a greatimprovement over the best systemic therapy.22,55,57-59 In ourinitial report, we used MMC as the agent,22 but havechanged to cisplatinum after the reports from the surgerybranch of the National Cancer Institute.57,58 We believethat mesothelioma represents one of the strongest cases forcombining IPHC with CS.

It is estimated that only a handful of patients who arepotential candidates for this therapy actually receive it,which is underscored by the relatively small number ofpatients accrued to the phase II studies for peritoneal car-cinomatosis at large perfusion centers. It is clear thatexpanding the number of centers should be done by surgi-cal oncologists who have more than a passing knowledge ofsystemic chemotherapy and are comfortable with the rigorsof aggressive operative procedures in the abdomen.23 Thishas led to a consensus statement by a group of surgeonswith an interest in CS and IPHC that outlines an evalua-tion strategy for PSD from colorectal carcinoma.52,54

Although reported results from perfusion centers rep-resent a substantial improvement in duration and likelyqualityof life,60-62 the majority of patients undergoing theseprocedures will experience tumor recurrence. Evaluatingpatients for second cytoreduction and additional che-moperfusion will become an ever more common problembecause patients with PMP often require multiple proce-dures.23,63We, and others, believe that in selected patients,a second cytoreductive procedure and chemoperfusionmay be of value. In evaluating patients for second cytore-

duction, the same criteria used to select patients for the firstremain important. Specifically, the patients must remainmedically fit to tolerate a major operative procedure, be freeof extraabdominal or hepatic parenchymal metastases, andhave disease that seems amenable to complete cytoreduc-

tion. Additionally, the time to recurrence after initial cy-toreduction and the completeness of the initial cytoreduc-tion should be considered in deciding to proceed withanother procedure. Patients with bulk residual disease afteran initial cytoreduction for PSD should not be consideredcandidates for second cytoreductive procedures.23,63 In thisstudy, 41 patients underwent a second (or third) IPHC.Although these patients had good outcomes, this clearlyrepresents a selection bias in choosing patients for repeatprocedures.

Several issues surround the future of CS and IPHC forPSD. Chief among them is how to make such therapy

standardized and available to large numbers of patients. Atpresent, there are approximately 25 active centers in the USand only 6 have experience with more than 100 patients.The operations required for aggressive cytoreduction arelengthy, challenging, morbid, and use a great deal of hos-pital, blood bank, and surgical house officer resources. Re-source use (and safety) of chemotherapy in the operatingroom is daunting for many centers. Additionally, great careneeds to be taken in selecting patients to undergo this pro-cedure. Our experience has evolved and improved over theyears of this study. This implies a learning curve, whichmay be lengthy. Although we would like to think our sur-

gical techniques have improved substantially over time, webelieve that it is primarily better patient selection that ac-counts for the improved outcomes. In addition, for mostprimary site tumors, the optimal time, dose, temperature,and chemotherapeutic agent for perfusion are not based onclass I data.

Fundamental questions about IPHC for PSD need to beaddressed. Foremost among these is whether the additionof IPHC after CS is of value. It seems obvious that thevalueof IPHC should depend on the tumor being treated. Theonly completed randomized trial for CS and IPHC evalu-ated patients with PSD from colorectal primary lesions.

That trial compared CS and IPHC with standard systemicchemotherapy to standard systemic therapy (fluorouraciland leucovorin) and found the CS and IPHC doubled thesurvival.9 But to date, no study has compared CS with orwithout IPHC.53 Clearly, it would be desirable to evaluatethe value of IPHC versus CS alone, in a multicenter, pro-spective randomized trial. But such a randomized con-trolled trial has proved difficult to complete and effortshave previously failed.53,64 Specifically, many patients pre-senting themselves for evaluation to specialized centers



9/11

refuse to consider such a randomization.64 Efforts to bringCS and IPHC to multicenter trials have not been embracedby the cooperative oncology groups to date. Such difficul-ties in performing randomized trials do not mean theyshould not be pursued.

The advancement of centers of excellence and the initi-ation of cooperative group trials will help define the opti-mal treatment approach for peritoneal spread for PSD.Thefuture of CS and IPHC for PSD lies in multicenter andrandomized trials that not only investigate response andsurvival, but also standardization of techniques, quality oflife, and integration with ever improving systemic therapy.

Author Contributions

Study conception and design: Levine, LoggieAcquisition of data: Levine, Stewart, Geisinger, Loggie,

Shen

Analysis and interpretation of data: Levine, Russell, Stewart,Loggie, ShenDrafting of manuscript: Levine, RussellCritical revision: Levine, Geisinger, Shen, Russell, Stewart

Acknowledgment: We wish to acknowledge the substantialcontributions of house officers, perfusionists, nurses, enteros-tomal therapists, psychosocial oncologists, urologists, and ra-diologists who contributed to the care of these patients. We

would also like to thank Joan Feder for her editorial supportand Mary Cromer for her ongoing database maintenance.

REFERENCES

1. Sadeghi B, Arvieux C, Glehen O, et al. Peritoneal carcinomato-sis from non-gynecologic malignancies: results of theEVOCAPE 1 multicentric prospective study. Cancer 2000;88:358363.

2. Sugarbaker PH. Peritonectomy procedures. Ann Surg 1995;221:2942.

3. Torisu M, Katano M, Kimura Y, et al. New approach to man-agement of malignant ascites with a streptococcal preparationOK 432: improvement of host immunity and prolongation ofsurvival. Surgery 1983;93:353357.

4. Hendren SK, Hahn SM, Spitz FR, et al. Phase II trial of debulk-ing surgery and photodynamic therapy for disseminated intra-peritoneal tumors. Ann Surg Oncol 2001;8:6571.

5. Sindelar WF, DeLaney TF, Tochner Z, et al. Technique of pho-todynamic therapy for disseminated intraperitoneal malignantneoplasms. Phase I study. Arch Surg 1991;126:318324.

6. Sugarbaker PH, Jablonski KA. Prognostic features of 51 colo-rectal and 130 appendiceal cancer patients with peritoneal car-cinomatosis treated by cytoreductive surgery and intraperitonealchemotherapy. Ann Surg 1995;221:124132.

7. Beaujard AC, Glehen O, Caillot JL, et al. Intraperitoneal che-mohyperthermia with mitomycin C for digestive tract cancerpatients with peritoneal carcinomatosis. Cancer 2000;88:25122519.

8. Loggie BW, Fleming RA, McQuellon RP, et al. Cytoreductivesurgery with intraperitoneal hyperthermic chemotherapy fordisseminated peritoneal cancer of gastrointestinal origin. AmSurg 2000;66:561568.

9. Verwaal VJ, van Ruth S, de Bree E, et al. Randomized trial ofcytoreduction and hyperthermic intraperitoneal chemotherapyversus systemic chemotherapy and palliative surgery in patients

with peritoneal carcinomatosis of colorectal cancer.J Clin Oncol2003;21:37373743.

10. Armstrong DK, Bundy B, Wenzel L, et al. Intraperitoneal cis-platin and paclitaxel in ovarian cancer. N Engl J Med 2006;354:3443.

11. Markman M, Walker JL. Intraperitoneal chemotherapy in ovar-ian cancer: A review, with a focus on practical aspects of treat-ment. J Clin Oncol 2006;24:988994.

12. Miner TJ, Shia J, Jaques DP, et al. Long-term survival followingtreatment of pseudomyxoma peritonei: an analysis of surgicaltherapy. Ann Surg 2005;241:300308.

13. Gough DB, Donohue JH, Schutt AJ, et al. Pseudomyxomaperitonei. Long-term patient survival with an aggressive regionalapproach. Ann Surg 1994;219:112119.

14. Hinson FL, Ambrose NS. Pseudomyxoma peritonei. Br J Surg1998;85:13321339.

15. Cannistra SA. Intraperitoneal chemotherapy comes of age.N Engl J Med 2006;354:7779.

16. Stewart JH, Shen P, Russell GB, et al. Appendiceal neoplasmswith peritoneal dissemination: outcomes after cytoreductivesurgery and intraperitoneal hyperthermic chemotherapy. AnnSurg Oncol 2006;13:111.

17. Bradley RF, Stewart JH, Russell G, et al. Pseudomyxoma peri-tonei of appendiceal origin: a clinicopathologic analysis of 101uniformly treated patients at a single institution, with literaturereview. Am J Surg Path 2006;30:551559.

18. Shen P, Levine EA, Hall J, et al. Factors predicting survival afterintraperitoneal hyperthermic chemotherapy with mitomycin C

after cytoreductive surgery for patients with peritoneal carcino-matosis. Arch Surg 2003;138:2633.19. Shen P, Hawksworth J, Lovato J, et al. Cytoreductive surgery

and intraperitoneal hyperthermic chemotherapy with mitomy-cin C for peritoneal carcinomatosis from nonappendiceal colo-rectal carcinoma. Ann Surg Oncol 2004;11:178186.

20. Hall JJ, Loggie BW, Shen P, et al. Cytoreductive surgery withintraperitoneal hyperthermic chemotherapy for advanced gas-tric cancer. J Gastrointest Surg 2004;8:454463.

21. Jacks SP, Hundley JC, Shen P, et al. Cytoreductive surgery andintraperitoneal hyperthermic chemotherapy for peritoneal car-cinomatosis from small bowel adenocarcinoma. J Surg Oncol2005;91:112117.

22. Loggie BW, Fleming RA, McQuellon RP, et al. Prospective trialfor the treatment of malignant peritoneal mesothelioma. AmSurg 2001;67:9991003.

23. Levine EA. Problems of success and problems of failure: recur-rent disease after cytoreductive surgery and intraperitoneal che-moperfusion. Ann Surg Oncol 2004;11:351353.

24. AJCC cancer staging manual, 6th ed. New York: Springer-Ver-lag; 2002.

25. Dedrick RL, Flessner MF. Pharmacokinetic problems in perito-neal drug administration: tissue penetration and surface expo-sure. J Nat Cancer Inst 1997;89:480487.

26. Dedrick RL. Interspecies scaling of regional drug delivery.J Pharm Sci 1986;75:10471052.

27. Flessner MF, Dedrick RL, Schultz JS. A distributed model of



10/11


11/11

vivorship and quality of life after cytoreductive surgery plusintraperitoneal hyperthermic chemotherapy for peritonealcarcinomatosis. Ann Surg Oncol 2003;10:155162.

62. Alexander HR, Mavroukakis SM, Libutti SK, et al. Impact oftumor resection and intraperitoneal chemotherapy on healthrelated quality of life in patients with peritoneal surface malig-nancies. Proceedings of the 57th Annual Society of Surgical

Oncology Cancer Symposium; 2004.63. Portilla AG, Sugarbaker PH, Chang D. Second-look surgery

after cytoreduction and intraperitoneal chemotherapy for peri-toneal carcinomatosis from colorectal cancer: analysis of prog-nostic features. World J Surg 1999;23:2329.

64. Elias D, Delperro JR, Sideris L, et al. Treatment of peritonealcarcinomatosis from colorectal cancer: Impact of complete cy-toreductive surgery and difficulties in conducting randomizedtrials. Ann Surg Oncol 2004;11:518521.

DiscussionKIRBY I BLAND, MD (Birmingham, AL): Dr Levine, it is verycommendable, your survival of 27%, which at least in my handsdoing this operation has been zero. My results would be in the 380

who had a zero 5-year survival rate.This is a very difficult procedure,and in your own terms was a 10-hour procedure that has been un-achieved by other surgeons/clinics.

You discussed differentiated histology types and broke out adeno-carcinoma, specifically of the GI tract versus other sites, such ascarcinoma of the appendix, having different survival outcomes.Knowing that data now, would you reconsider patients who wouldnot be inclusive for the extensive, protracted (and expensive) proce-

dure you describe? In other words, would you exclude those patientswith adenocarcinoma of the GI tract even if they did have negativeparameters outside the abdomen? And have you considered submit-ting this technique in a prospective randomized analysis, such as the

American College of Surgeons OncologyTrials Group, such that thiscould truly be studied objectively?

KELLY M MCMASTERS, MD (Louisville, KY): The authors re-port their experience with over 500 cases of cytoreductive surgery

with intraperitoneal hyperthermic chemotherapy for a variety of ma-lignancies over the past 15 years. The procedure was associated witha 43% complication rate and a 30-day mortality rate of 4.8%or4.3% in the updated analysis. Clearly, this is a serious operation withmorbidity and mortality rates as high as or higher than any electiveoperation we perform in surgical oncology. Median survival wasabout 22 months, with 5-year survival rate of 28%.

My questions are, first, does this operation really work? And sec-ond, if so, who really benefits from it? I would like to ask a couple ofquestions related to number two first, and then get to question num-ber one.

Some patients may have disease that is very indolent and may notrequire this procedure at the time of diagnosis. Others may havebulky or more aggressive tumors with a poor performance status, andtheir chance of benefit from the procedure is small, with a concom-itant high likelihood of bad surgical outcomes. Clearly, performance

status, type of primary tumor, disease burden, and development of

complications all had a major impact on survival. So my question is:

How have you used the data from this study to modify your selection

criteria? Will you only perfuse patients with ECOG performance

status of zero to one? Did you perform analyses to predict which

patients are likely to have major complications and early postopera-

tive death?Finally, concerning question number one: Does this operation

really work? I am afraid this paper does little to answer that question.

A single randomized study suggests it is effective for patients with

carcinomatosis from colorectal cancer. Because randomized trials

have been so elusive for this procedure, my final question for the

authors is: Are you prepared to restrict your practice of intraperi-

toneal hypothermic chemotherapy to properly conducted phase

III studies? That is the only way we will ever answer this question.

If even a modest fraction of the patients in this analysis had been

enrolled in a prospective randomized study, we would now have

convincing evidence to support its use. Therefore, I challenge the

authors and those relatively few other centers in the US thatperform this procedure to join together and answer this question

once and for all.

JOHN M DALY, MD (Philadelphia, PA): This particular report is of

great personal interest to me. Your results, Dr Levine, and those of

Paul Sugarbaker, David Bartlett, and a number of others, are really

quite similar.

It is curious that among our gynecologic oncology colleagues this

treatment method, that is, hyperthermic intraperitoneal chemother-

apy, is not uniformly accepted, and yet they are treating a more

favorabledisease,ovarianprimary cancer. I have a coupleof questions

for you.

One is the anastomosis that is performed at the time of the treat-ment. Other centers have reported a very large breakdown rate of the

anastomosis and subsequent fistula, infection, and mortality. Could

you tell us a little bit more about how you manage patients with

bowel resection at the time of intraperitoneal treatment?

Second, you mentioned some patients undergo a second-lookpro-

cedure. Could you describe a little bit about that, the adhesions that

are found at the time of second-look operation? Some have suggested

that the type of chemotherapy, whether it be mitomycin or cisplatin,

might be different in terms of subsequent adhesion formation.

Finally, I couldnt agreemore with theissue of a prospective trial to

evaluate this form of treatment, although there are only a few centers

throughout the country that are doing this. Without such a trial we

will never know the answers to some of the questions you posed.Perhaps you could tell us how you would design the trial that would

be used to answer the question.

EMMANUEL ZERVOS, MD (Tampa, FL): I would like to congrat-

ulate Dr Levine and colleagues. We are kind of a renegade shop in

Tampa doing this, and many times we utilize your data to justify the

operations that we undertake. I just have a couple questions for you.

There has been recent data from Europe and Dr Elias using oxali-

platium and the coliseum technique rather than the closed tech-

nique, which has really quite remarkable results. And I wonder if you


Cytoreductive Surgery and Intraperitoneal

Documents

Transcript of Cytoreductive Surgery and Intraperitoneal