Diagnosis and Treatment of Peritoneal Mesothelioma: The Washington Cancer Institute Experience

Paul H. Sugarbaker, Yair I.Z. Acherman, Santiago Gonzalez-Moreno, Gloria Ortega-Perez, O. Anthony Stuart, Pierre Marchettini, and Dal Yoo

(From the Washington Cancer Institute, Washington, DC. Address reprint requests to Paul H. Sugarbaker, MD, Washington CancerInstitute, 110 Irving St, NW, Washington, DC 20010. Copyright © 2002 by WB. Saunders Company 0093-7754/02/2901-0012$35.00/0 doi:10.10531sonc.2002.30236)

Peritoneal mesothelioma is a rare disease, but increasing in frequency. The incidence is approximately one per 1,000,000 and about one fifth to one third of all mesotheliomas are peritoneal. Because of its unusual nature, the disease has not been clearly defined either in terms of its natural history, diagnosis, or management. This article reviews a single institution's experience with 51 patients prospectively treated over the past decade with increasingly aggressive local/regional protocols. Peritoneal mesothelioma patients generally present with two types of symptoms and signs; those with abdominal pain, usually localized and related to a dominant tumor mass with little or no ascites and those without abdominal pain, but with ascites and abdominal distention. Pathologically, a positive immunostain for calretinin has markedly increased the accuracy of diagnosis. Prognosis as determined by clinical presentation, the completeness of cytoreduction, and gender (females survive longer than males) appears to be improved by the use of intraperitoneal chemotherapy. Over the past decade, the management of these patients has evolved similarly to ovarian cancer treatment and now involves cytoreductive surgery, heated intraoperative intraperitoneal chemotherapy (HIIC) with cisplatin and doxorubicin, and early postoperative intraperitoneal paclitaxel. These perioperative treatments are followed by adjuvant intraperitoneal paclitaxel and second-look cytoreduction. Prolonged disease-free survival and reduced adverse symptoms with the current management strategy are documented by a high complete response rate as assessed by a negative second-look. This multimodality treatment approach with cytoreductive surgery and intraperitoneal chemo-therapy has resulted in a median survival of 50 to 60 months. Peritoneal mesothelioma is an orphan disease that is treatable with expectations for "potential" cure in a small number of patients if diagnosed and treated early with definitive local/regional treatments. A prolonged high quality of life is possible in the majority of patients.
Semin Oncol 29.51-61. Copyright © 2002 by WB. Saunders Company.

MESOTHELIOMA is a neoplasm originating from the mesothelial cells lining the human body cavities. Mesothelioma may involve the pleura, less frequently the peritoneum, and, rarely, the pericardium and the tunica vaginalis testis. Peritoneal mesothelioma is usually a rapidly fatal peritoneal surface malignancy with a median survival of less than 1 year. It represents about one fifth to one third of all forms of mesothelioma. Asbestos exposure appears to be causative in some cases of peritoneal mesothelioma, but a search for other carcinogens continues. This report describes our experience with 51 patients with peritoneal mesothelioma, and analyzes their symptoms, clinicopathologic features, and survival after aggressive local-regional treatment.

DEMOGRAPHIC AND EPIDEMIOLOGIC DATA
The mean age for these patients was 53 years, with a range of 16 to 78 years. Forty-nine patients were Caucasian; there was one African American and one Asian American. Sixteen patients had no history of exposure to asbestos, 20 had a positive history, and no data are currently available an 15 patients.
Nineteen patients had a family history of cancer in a parent or sibling and five had more than one first-degree relative with a malignancy; no data are available an nine patients (Table 1).
Additional epidemiologic studies of this group of peritoneal mesothelioma patients are ongoing.

DIAGNOSIS OF PERITONEAL MESOTHELIOMA
The initial symptom that led to a diagnosis of peritoneal mesothelioma was prospectively recorded in all 51 patients. In the past, it was assumed that the initial symptom in virtually all patients was an expanding abdomen from malignant ascites. However, as shown in Table 2, these patients can present with a variety of symptoms.
We categorized the patients into three groups based on presenting symptoms. Approximately one third present with abdominal distention, one third with abdominal pain, usually localized, and the remaining third present with combined symptoms of distention, pain, and other findings. We have designated these three types as a "wet type" of peritoneal mesothelioma presenting with Symptoms of malignant ascites causing increased abdominal girth, a "dry-painful type" presenting with a focal mass seen an computed tomography (CT) scan causing pain, and a "combined type" characterized by both pain and ascites. We attempted to correlate the histologic findings with patient symptoms at presentation but no significant relationships were apparent.

Patients presenting with an acute abdomen with peritoneal mesothelioma have not previously been described. Four patients had the sudden onset of severe pain requiring evaluation at an emergency treatment facility; three had acute appendicitis, and one of these three had a perforated appendix.
A single patient presented with an incarcerated umbilical hernia. In addition to pain, increased abdominal girth and new-onset hernia, signs and Symptoms included weight loss, infertility, shortness of breath, fever, and night sweats.

MECHANISMS OF DEFINITIVE DIAGNOSIS OF MESOTHELIOMA
Clinically, the vague nature of the Symptoms and signs, the long-term presence of these Symptoms and signs prior to definitive diagnosis, and the young age of many of these patients led to a delay in diagnosis. Laparoscopy was the most common diagnostic test required for definitively diagnosing the wet type of mesothelioma (64%). Cytology of fluid removed by paracentesis rarely resulted in a definitive diagnosis.
An important caveat must accompany the recommendation for laparoscopy in the diagnosis of peritoneal mesothelioma. In out follow-up, port site recurrence is observed at nearly all trochar sites. Our recommendation is to limit the puncture sites to along the linea alba. As seen in the CT scan shown in Fig 1, cytoreductive surgery plus intraperitoneal chemotherapy may stabilize the peritoneal surface disease while trochar site disease may progress rapidly.

A unique group of patients were the four wich the "benign cystic mesothelioma." These patients all presented with abdominal pain. Several patients were on large doses of narcotics prior to their definitive treatment. Three of the four had been labeled as malingerers as a result of their numerous visits to emergency departments for complaints of abdominal pain. Because of the noninvasive character of the disease and its definitive treatment by peritonectomy procedures plus intraperitoneal chemotherapy, a curative result with return to full health is expected in this highly symptomatic group of patients. One of these patients, after a 10-year delay in definitive treatment, showed a malignant transformation from benign cystic mesothelioma to an invasive (malignant) mesothelioma.

CT in the Diagnosis of Peritoneal Mesothelioma
The two clinical types of peritoneal mesothelioma, wet or dry-painful type, have profoundly different appearances by CT examination. In the wet type, there is little or no evidence of solid tumor.

Fig 1. CT of abdomen in a patient who had lateral trochar sites in a laparoscopy to diagnose peritoneal mesothelioma.

Occasionally, small nodules lining the parietal peritoneal surfaces are evident, especially beneath the right hemidiaphragm (Fig 2). The CT/radiologic presentation of the dry-painful type of peritoneal mesothelioma may disclose several mass lesions, but often there is a dominant mass isolated to one part of the abdomen. Likewise, the tumor mass is commonly associated with the greater omentum. Usually the symptomatic mass lesion seen an CT is thought to be an intra-abdominal abscess or a large primary adenocarcinoma. Only at laparotomy is the definitive diagnosis evident in these patients presenting wich solid tumor in the absence of ascites (Fig 3).

Fig 2. CT of wet type of peritoneal mesothelioma. Fluid conforms to the shape of the small bowel and its mesentery. A small volume of solid tumor is layered out an the parietal peritoneum.

The preoperative CT is of great value in predicting the benefits of cytoreductive surgery plus intraperitoneal chemotherapy. Complete removal of all visible tumor using peritonectomy procedures is often predicted by the CT. To facilitate successful surgery, the Small bowel and its mesentery should be compartmentalized apart from the malignancy and there should be no tumor nodules directly adjacent to the small bowel. No disruption of intestinal function should be present, ie, segmental bowel obstruction. If these criteria of (1) compartmentalization and (2) no evidence of obstruction are observed an CT scan, then marked palliative and possible curative efforts from cytoreductive surgery with intraperitoneal chemotherapy are expected. Cystic peritoneal mesothelioma will commonly show these CT findings (Fig 4).

Fig 3. CT of dry painful type of peritoneal mesothelioma. Discrete masses are present at multiple sites around the abdomen. These masses are painful to deep palpation.

One of the greatest difficulties encountered in the management of patients with peritoneal mesothelioma involves prior and repeated debulking procedures. In this situation the malignant cells become uniformly distributed over the peritoneal surfaces including small bowel and small bowel mesentery. This leads to solid tumor diffusely distributed in large volumes and intimately attached to the small bowel. This results in a hopeless situation for further surgery (Fig 5). In this condition, the patient becomes untreatable by localregional treatment strategies. Occasionally, aggressive systemic chemotherapy results in a transient response, but no long-term survival occurs once fixation of the bowel has taken place.

Fig 4. CT of cystic type of peritoneal mesothelioma. Thin walled cysts filled with clear fluid fill the abdominal and pelvic spaces. The Small bowel is pushed to the side (compartmentalized). This separation of viscera from tumor is associated with a good prognosis with cytoreductive surgery plus intraperitoneal chemotherapy.

Pathology of Peritoneal Mesothelioma
The correct pathological diagnosis of peritoneal mesothelioma is necessary in that a variety of other abdominal and pelvic malignancies will present with peritoneal seeding. Approximately 10% of patients with primary colon cancer will have peritoneal carcinomatosis. Up to 30% of patients with gastric cancer and pancreas cancer have peritoneal seeding at the time of exploration for resection of the primary malignancy. A majority of patients with papillary serous ovarian cancer have peritoneal seeding. The simultaneous occurrence of one of these common cancces coincidentally with a primary peritoneal mesothelioma can occur. Interestingly, one patient in our series who was diagnosed with peritoneal carcinomatosis of colorectal origin subsequently presented 2 years later with peritoneal mesothelioma.

Fig 5. CT of peritoneal mesothelioma after multiple debulking procedures. The small bowel is fixed by tumor. Further surgical interventions are not possible.

The histopathologic examination by hematoxylin and eosin stain remains an important part of the diagnosis of peritoneal mesothelioma. Cords of cuboidal and polygonal epithelial cells within a mucinous matrix are prominent. A papillary growth pattern that emerges within areas of apparent mesothelial hyperplasia are often evident (Fig 6).

Fig 6. Epithelial mesothelioma. High-power view of cuboidal or low columnar tumor cells, with acidophilic cytoplasm and relative uniform vesicular nuclei with prominent nucleoli (hematoxylin and eosin; original , magnification x 500).

The architectural features of the peritoneal mesothelioma may be quite heterogeneous including solid, trabecular, cord-like, focal glandular, and papillary appearances.
Table 3 summarizes the immunohistochemical (IHC) stains of adenocarcinoma and peritoneal mesothelioma. Figure 6 shows that the calretinin immunostain may definitively identify the disease. Simultaneous staining of tissues with 1372.3 (negative stain) and calretinin (positive stain) is nearly a definitive pattern, but it must be emphasized that no single IHC stain can make the final diagnosis of mesothelioma versus carcinomatosis (Fig 7).

Fig 7. Epithelial mesothelioma. High magnification of tumor cells with strong positive expression for calretinin (calretinin; original magnification x 500).

Pathologists at our institution have identified seven different pathologic types of peritoneal mesothelioma among the 51 patients treated (Table 4).

Fig 8. Pharmacokinetics of intraperitoneal paclitaxel at 20 mg/m2 in 1 L of peritoneal dialysis solution. In this patient the chemotherapy solution was instilled as rapidly as possible an the first postoperative day.

The morphologic type did not correlate with prognosis but surgical selection and referral bias may skew these findings. In the past, electron microscopy was important in the definitive pathologic diagnosis of mesothelioma. Today, the sophisticated use of an extensive panel of IHC markers can be used to confirm the diagnosis of mesothelioma.

TREATMENTS
Rationale for Aggressive Local-Regional Approach
Peritoneal mesothelioma is a disease that produces symptoms and causes the patients death as a result of disease progression confined to the surfaces of the abdomen and pelvic. Sometimes peritoneal mesothelioma will penetrate through the right or left hemidiaphragm and involve the pleura, leading to respiratory symptoms. However, in a majority of patients, metastases to other sites do not occur. Death is usually caused by intestinal obstruction, fistula formation, and resulting starvation.
In the dry-painful type of mesothelioma, multiple sites of cancer progression may be evident, although one clinically dominant mass may occur.
Characteristically, however, patients in the earliest stages of the disease Show a multifocal disease process characterized by the simultaneous involvement of the various peritoneal surfaces. The disease is hypothesized to be a field defect within the abdomen and pelvis. A single site for disease that spreads throughout the abdomen and pelvis as in carcinomatosis does not seem to occur. The diffuse nature of the disease within the abdomen and pelvis, its Symptoms, and the disease progression causing obstruction of the gastrointestinal tract has led our group to pursue aggressive local-regional treatment strategies for selected patients with peritoneal mesothelioma. Over the course of a decade, an increasingly aggressive approach has evolved. Peritonectomy procedures are combined with intraoperative, early postoperative, and delayed cycles of intraperitoneal chemotherapy. In fit patients who are responding well to therapy, a second-look surgery is often recommended.
A key to these aggressive local-regional treatment strategies is the use of an intraperitoneal route of chemotherapy in the perioperative period after the surgeon has maximally cytoreduced the mesothelioma. Maximal contact and penetration of the chemotherapy into residual cancer should be possible at this time, since there are no adherences restricting the uniform distribution of chemotherapy to all of the peritoneal surfaces.
Recent pharmacologic studies in humans with paclitaxel show that it is an ideal drug for intraperitoneal use. As shown in Fig 8, there is a marked sequestration of the drug within the peritoneal space an the first to fifth postoperative days even though extensive peritonectomy has been performed. Pharmacologic studies of the intraperitoneal and plasma concentrations of paclitaxel after intraperitoneal administration at 20 mg/m2 show 2,000 times the drug concentration in the peritoneal cavity compared to the serum levels.
The goal of treatment is to use both surgery and regional chemotherapy maximally in every patient, to the limits of patient tolerance. The 33 patients previously reported had a variety of treatments. Nine patients with extensive ascites were treated before surgery with induction intraperitoneal chemotherapy using cisplatin (Platinol, Bristol Myers Squibb, Pennington, NJ) 15 mg/m2/d and doxorubicin (Adriamycin, BenVenue Laboratories, Bedford, OH) 6 mg/m2/d. Each cycle of treatment was for 5 consecutive days, given once a month for 3 months. After induction intraperitoneal chemotherapy, the patients went an to cytoreductive surgery. In the other 24 patients, the initial treatment at this institution was cytoreductive surgery, which included all or part of the following five peritonectomy procedures, which have been described previously: greater and lesser omentectomy, right and left subphrenic peritonectomy, and complete pelvic peritonectomy. Visceral resections were often required for complete cytoreduction. During the surgical procedure the patients had intraoperative intraperitoneal chemotherapy with cisplatin 50 mg/m2 and doxorubicin 15 mg/m2, heated to 41 to 41.5°C. During the 90 minutes of chemotherapy, additional debridement of cancer from the small bowel and its mesentery occur. The heated chemotherapy is manually mixed by the surgeon to assure uniform distribution.
Recently, early postoperative intraperitoneal paclitaxel (20 mg/m2/d X 5) has been added to the perioperative treatments. Also, intraperitoneal paclitaxel given by a permanent intraperitoneal port has been used as an adjuvant for 6 months postoperatively. Second-look surgery and a final cycle of heated intraoperative intraperitoneal chemotherapy completes the aggressive local-regional approach.

TREATMENT DATA
Between 1989 and 1999,47 cytoreductive procedures were performed an 33 patients with peritoneal mesothelioma by the same surgeon: 33 as a first procedure, 11 as a second-look, and three as a theed-look surgery. Data regarding those 33 procedures are summarized in Table 5. Twenty patients received a major first cytoreduction involving four or more abdomino-pelvic regions. Thirteen patients, with high comorbidity or unresectable peritoneal mesothelioma, had a less extensive debulking procedure; survival approached statistical significance in favor of major cytoreduction (P = .07). Mean operating time was 6.8 hours (range, 2 to 13.5 hours); mean ascites volume was 2.4 L (range, 0 to 14 L); mean blood requirement was 2 units (range, 0 to 8 units). Twenty-eight patients received perioperative intraperitoneal chemotherapy (heated intraoperative treatment, postoperative treatment or both) and five patients did not; their median survival durations were 28 and 18 months, respectively (P = .11) Eleven patients underwent a scheduled second-look cytoreduction 6 to 9 months after the first operation, combined with perioperative intraperitoneal chemotherapy in 10. Three patients had a theed-look operation. The median survival duration was 35 months in the 11 patients who had a second-look surgery compared with 16 months in the other 22 patients; this was statistically significant (P =.002). The survival advantage in favor of second look surgery may have resulted from selection factors inherent to that group of patients.

We have attempted to refine a quantitative system for this type of surgery. For the initial 33 cytoreductions, the mean peritoneal cancer index was 29. The peritoneal cancer index is a score compiled by the surgeon as the abdomen is explored and all adhesions are lysed. The peritoneal cavity is divided into nine regions by two equally spaced horizontal and vertical lines. An additional four regions are defined by upper and lower jejunum, and upper and lower ileum. The volume of cancer distributed in each of these 13 abdominopelvic regions is scored as 0 (no cancer seen), 1 (nodules < 0.5 cm), 2 (nodules 0.5 to 5 cm), and 3 (nodules > 5 cm). The maximum peritoneal cancer index would be 39 (13 X 3). For patients with a score of 0 to 28, the median survival duration was 34 months compared with only 16 months for patients with a score between 29 and 39; the peritoneal cancer index was a statistically significant predictor of survival (P = .026).
The completeness of cytoreduction score was another quantitative prognostic indicator that was a statistically significant predictor of survival. The score was determined by the surgeon at the time of completion of the cytoreductive surgery. A score of zero indicated that no cancer nodules were seen. One indicated tumor nodules less than 0.25 cm. Two indicated tumor nodules 0.25 to 2.5 cm. Three indicated tumor nodules of greater than 2.5 cm or a confluence of cancer nodules at any site.16 The completeness of cytoreduction score was 0 to 2 in 17 patients and 3 in 16 patients, with median survival durations of 41 and 13 months, respectively (P = .0002).
The median survival of our 51 patients is shown in Fig 9. There are Jong-term survivors over 5 years following these treatments. All patients treated at the Washington Cancer Institute within the evolution of treatment strategies outlined above are included in these data. The challenge for our group at this point is to eliminate from these aggressive local-regional treatment strategies those patients who rapidly fail to respond. The performance status of the patient, the extent of disease as estimated by the peritoneal cancer index, the completeness of cancer resection, the clinical type (wet v dry-painful), and gender all seem to be important.

MORBIDITY AND MORTALITY
From our recent report of 33 patients with peritoneal mesothelioma, there were 11 patients with grade III to IV complications, giving an overall morbidity rate of 33%. Four patients required reoperation: one for persistent bile leak from the liver surface, one for small bowel fistula, and two for late intraabdominal bleeding. There were three perioperative deaths (6%) from sepsis that occurred after surgery.

Fig 9. Kaplan-Meier survival curve of our 51 patients with peritoneal mesothelioma. Median survival from diagnosis was 56 months.

DISCUSSION
Peritoneal mesothelioma has emerged in surgical thinking over almost 100 years from a clinical oddity to an established rare disease of increasing importance. In 1908 Miller and Wynn17 first described "...a malignant tumor arising from the endothelium of the peritoneum and producing mucoid ascitic fluid." In 1909 Adami1S first used the term "mesothelioma" to refer to a highly malignant tumor of the serosal membranes involving mostly the pleura and the peritoneum, and less often the pericardium and tunica vaginalis testis. Mesothelioma is not specific to humans and is also found in other mammalian species, including horse, monkey, dog, and cow.
Peritoneal mesothelioma represents 20% to 37% of all mesotheliomasl-3; because of its rarity and Small size of published series, the best estimate of annual incidence in the United States is 200 to 400 new cases. In a review of the literature, no dominant therapeutic guidelines for peritoneal mesothelioma are found. Most articles are clinicopathological retrospective reviews or case reports compiling disparate therapeutic experiences, including the use of systemic chemotherapy, whole abdominal radiation, and intraperitoneal treatments with compounds such as colloidal radioactive "P and 19"Au, thiotepa, and bleomycin. Recent reports show a more systematic approach to peritoneal mesothelioma, with debulking surgery and systemic chemotherapy including paclitaxel, cisplatin, or doxorubicin. A phase I trial with 18 patients treated at the National Cancer Institute used an approach similar to ours with promising results.
In spite of these approaches, the prognosis is grim after diagnosis of peritoneal mesothelioma. The median survival reported in the literature is 7 to 13.5 months and only a few long-term survivors are reported. The median survival duration in our patients was 56 months from diagnosis, with a projected 3-year survival rate of approximately 60%. These results need to be interpreted with caution due to short follow-up and a selected study population. In our experience, the positive predictive factors for survival identified by univariate analysis included: (1) female gender, (2) good health status, (3) presentation in the absence of increased abdominal girth, (4) low peritoneal cancer index, (5) complete cytoreduction, and (6) second-look surgery (for selected patients). There was a suggestion that prior surgery (partial debulking or resection of mesothelioma) in the absence of intraperitoneal chemotherapy may jeopardize subsequent treatment and survival. The peritoneal surface malignancy may move more rapidly to an invasive process if anatomical defenses are disrupted by surgery or by laparoscopy. In patients who had trochar-site tumor progression, the abdominal wall masses seemed to progress more rapidly than residual tumor in the abdomen.
Longer survival was seen in patients in whom major resection was possible, (reflected by a good completeness of cytoreduction score). Patients with minimal or no progression of disease and no systemic metastases were candidates for a secondlook reoperation. In this selected group, the survival was prolonged (P = .019). These findings suggest, but do not confirm, that aggressive cytoreductive surgery is of value in this group of patients.
There was no statistical indication that intraperitoneal chemotherapy resulted in survival benefit. It is possible that selection factors were important in this evaluation, because patients with a large volume of mesothelioma after cytoreduction were usually given intraperitoneal chemotherapy. However, it may be unwise to conclude that intraperitoneal chemotherapy may be eliminated from the treatment regimen. Compared with previous experience with this disease, the benefits of intraperitoneal chemotherapy in palliation were thought to be substantial. Palliation of ascites may be an important outcome with intraperitoneal chemotherapy. All but one patient had no further symptoms from ascites after cytoreduction and perioperative intraperitoneal chemotherapy. In the group of patients with resection of all gross disease, intraperitoneal chemotherapy may have been an essential component of the treatment that resulted in long-term disease stabilization. With the current state of knowledge, a recommendation for intraperitoneal cisplatin, doxorubicin, and paclitaxel as a treatment for peritoneal mesothelioma, combined with cytoreductive surgery, may be worthy of consideration.

REFERENCES
1. Sridhar KS, Doria R, Raub WA Jr, et al: New strategies are needed in diffuse malignant mesothelioma. Cancer 70: 2969-2979, 1992

2. Antman K, Shemin R, Ryan L, et al: Malignant mesothelioma: prognostic variables in a registry of 180 patients, the Dana-Farber Cancer Institute and Brigham and Women's Hospital experience over two decades, 1965-1985. J Clin Oncol 6:147-153, 1988

3. Asensio JA, Goldblatt P, Thomford NR: Primary malignant peritoneal mesothelioma. A report of seven cases and a review of die literature. Arch Surg 125:477-481, 1990

4. Wagner JC, Sleegs CA, Marchand P: Diffuse pleural mesothelioma and asbestos exposure in North Western Cape Province. Br J Ind Mag 17:260-271, 1960

5. Peterson JT Jr, Greenberg SD, Buffler P: Non-asbestosrelated malignant mesothelioma. A review. Cancer 54:951960, 1984

6. Acherman YIZ, Welch L, Sugarbaker PH: Follow-back epidemiology of peritoneal mesothelioma. (submitted)

7. Jacquet P, Jelinek JS, Sugarbaker PH: Abdominal computed tomographic scan in the selection of patients with mucinous peritoneal carcinomatosis for cytoreductive surgery. J Am College Surg 181:530-538, 1995

8. Alcorn KW, Yan H, Shmookler BM, et al: Differential diagnosis of the simultaneous occurrence of colon carcinoma and peritoneal mesothelioma by immunohistochemistry. Surg Rounds 23:411-417, 2000

9. Sugarbaker PH, Yan H, Grazi R, et al: Early localized peritoneal mesothelioma as an incidental finding at laparoscopy: Report of a case and implications regarding natural history of the disease. Cancer 89:1279-1284, 2000

10. Markman M, Rowinsky E, Hakes T, et al: Phase 1 trial of intraperitoneal taxol: A Gynecologic Oncology Group study. J Clin Oncol 10:1485-1491, 1992

11. Markman M, Francis P, Rowinsky E, et al: Intraperitoneal Taxol (paclitaxel) in the management of ovarian cancer. Arm Oncol 5:S55-858, 1994 (suppl 6) PERITONEAL MESOTHELIOMA

12. Averbach AM, Sugarbaker PH: Peritoneal mesothelioma treatment approach based an natural history. Cancer Treat Res 81:193-211, 1996

13. Sebbag G, Yan H, Shmookler BM, et al: Results of treatment of 33 patients with peritoneal mesothelioma. Br J Surg 87:1-7, 2000

14. Sugarbaker PH: Peritonectomy procedures. Ann Surg 221:29-42, 1995

15. Stephens AD, Belliveau JF, Sugarbaker PH: Intraoperative hyperthermic lavage with cisplatin for peritoneal carcinomatosis and sarcomatosis, in Sugarbaker PH (ed): Peritoneal Carcinomatosis: Drugs and Diseases. Boston, MA, Kluwer Academic, 1996, pp 15-30

16. Jacquet P, Sugarbaker PH: Current methodologies for clinical assessment of patients with peritoneal carcinomatosis. J Exp Clin Cancer Res 15:49-58, 1996

17. Miller J, Wynn WH: A malignant tumor arising from the endothelium of the peritoneum, and producing mucoid ascitic fluid. J Pathol Bacteriol 12:267-78, 1908

18. Whitaker D, Manning LS, Robinson BWS, et al: The pathobiology of the mesothelium, in Henderson DW, Shilkin KB, Langlois SLP, et al (eds): Malignant Mesothelioma. New York, NY, Hemisphere Publishing, 1992, pp 25-68

19. Eltabbakh GH, Piver MS, Hempling RH, et al: Clinical picture, response to therapy, and survival of women with diffuse malignant peritoneal mesothelioma. J Surg Oncol 70:6-12, 1999

20. Markman M, Kelsen D: Efficacy of cisplatin-based intraperitoneal chemotherapy as treatment of malignant peritoneal mesothelioma. J Cancer Res Clin Oncol 118:547-550, 1992

21. Park BJ, Alexander HR, Libutti SK, et al: Treatment of primary peritoneal mesothelioma by continuous hyperthermic peritoneal perfusion (CHPP). Ann Surg Oncol 6:582-590, 1999