ORIGINAL ARTICLE – GASTROINTESTINAL ONCOLOGY

Diffuse Malignant Peritoneal Mesothelioma: Failure AnalysisFollowing Cytoreduction and Hyperthermic IntraperitonealChemotherapy (HIPEC)

D. Baratti, MD1, S. Kusamura, MD, PhD1, A. D. Cabras, MD2, P. Dileo, MD3, B. Laterza, MD1,

and M. Deraco, MD1,4

1Department of Surgery, National Cancer Institute, Milan, Italy; 2Department of Pathology, National Cancer Institute,

Milan, Italy; 3Department of Oncology, National Cancer Institute, Milan, Italy; 4Fondazione IRCCS, Istituto Nazionale

Tumori Milano, Milan, Italy

ABSTRACT Improved survival has been reported for

diffuse malignant peritoneal mesothelioma (DMPM) treated

by cytoreduction and hyperthermic intraperitoneal chemo-

therapy (HIPEC). The issue of treatment failure has never

been extensively addressed. The present study assessed the

failure pattern, management, and outcome of progressive

DMPM following comprehensive treatment. Clinical data

on 70 patients with DMPM undergoing cytoreduction and

HIPEC were prospectively collected; after a median follow-

up of 43 months, disease progression occurred in 38

patients. Progressive disease distribution in 13 abdomino-

pelvic regions was analyzed. In 28 patients undergoing

adequate cytoreduction (residual tumor B2.5 mm), clini-

copathological factors correlating to disease progression in

each region were investigated. Median time to progression

was 9 months [95% confidence interval (CI) 1.6–35.9].

Median survival from progression was 8 months (95% CI 4–

16.2). The failure pattern was categorized as peritoneal

progression (n = 31), liver metastases (n = 1), abdominal

lymph-node involvement (n = 2), pleural seeding (n = 4).

Small bowel was the single site most commonly involved

(n = 27). Residual tumor B2.5 mm (versus no visible) was

the only independent risk factor for disease progression in

epigastric region (P = 0.047), upper ileum (P = 0.029),

upper jejunum (P = 0.034), and lower jejunum (P =

0.002). Progressive disease was treated with second HIPEC

in 3 patients, debulking in 4, systemic chemotherapy in 16,

and supportive care in 15. At multivariate analysis, time to

progression \9 months (P = 0.009), poor performance

status (P = 0.005), and supportive care (P = 0.003) corre-

lated to reduced survival from progression. We conclude

that minimal residual disease, compared with macroscopi-

cally complete cytoreduction, correlated to failure in critical

anatomical areas, suggesting the need for maximal cytore-

ductive surgical efforts. In selected patients, aggressive

management of progressive disease seems worthwhile.

Once regarded as a rare and uniformly lethal disease,

diffuse malignant peritoneal mesothelioma (DMPM) is

attracting growing scientific interest.1 Analogously to

pleural mesothelioma, the incidence of the disease is stea-

dily increasing in developed countries, presumably due to

the widespread environmental and occupational exposure to

asbestos over the last decades.2,3 Also, genetic and molec-

ular knowledge of DMPM has considerably increased in

recent years, as have the treatment options.3–6 Specifically,

the comprehensive approach pioneered by Sugarbaker

involving surgical cytoreduction and heated perioperative

local-regional chemotherapy has reportedly resulted in a

median survival of 52–92 months, as compared with

9–12 months in historical case series treated by palliative

surgery and systemic or intraperitoneal chemotherapy.7–15

Despite encouraging results, approximately 40–60% of

patients develop disease progression and die of DMPM

following comprehensive treatment.7–12 In this context,

considerable efforts have been made to investigate the

selection factors predicting poor prognosis.9,11,12 However,

data on patients who fail to respond to initial treatment are

lacking and optimal management of recurrent or progressive

DMPM has never been defined. The analysis of the patterns

of failure is a different approach. It aims at understanding

� Society of Surgical Oncology 2008

First Received: 9 May 2008;

Published Online: 12 December 2008

M. Deraco, MD

e-mail: marcello.deraco@istitutotumori.mi.it

Ann Surg Oncol (2009) 16:463–472

DOI 10.1245/s10434-008-0219-1

how and possibly why a given treatment failed, in order to

identify the treatment modification which might improve the

clinical results. In 1996, a series of 32 patients with pseu-

domyxoma peritonei recurring after adequate cytoreduction

and early postoperative intraperitoneal chemotherapy

greatly contributed to overcome the inadequacies of such a

treatment strategy with the adoption of heated intraoperative

chemotherapy.16 More recently, exhaustive failure analyses

in patients with carcinomatosis of appendiceal and colo-

rectal origin have been reported by The Netherlands and the

Washington Cancer Institute.17–20

In the present study, we analyzed a prospective database

of patients with DMPM undergoing cytoreduction and

hyperthermic intraperitoneal chemotherapy (HIPEC) in an

attempt to improve the management of this disease. In detail,

primary study end-points were to assess: (1) the pattern of

failure following combined treatment, (2) the potential

clinicopathological factors determining treatment failure,

and (3) the management and outcome of progressive DMPM.

PATIENTS AND METHODS

All the patients included in the present study were

treated according to a protocol approved by the Institu-

tional Ethics Committee and signed written informed

consent. Data for this analysis derived from a prospective

database. Additional information was collected from the

medical records.

From January 1996 to April 2008, 70 consecutive patients

with DMPM were treated with cytoreduction and HIPEC by

the same surgical team at the National Cancer Institute

(Milan, Italy). Patients with multicystic or papillary well-

differentiated mesothelioma were excluded, as they repre-

sent different biological disease entities. One patient who

suffered postoperative death and two who died from other

causes were excluded from further analysis. After a median

follow-up of 43 months (range 1–129 months), disease

progression occurred in 38 patients, who constituted the

present study population.

Primary Cytoreduction and HIPEC

Diagnosis of DMPM was made or confirmed in our

Pathology Department according to a previously described

protocol including hematoxylin–eosin-stained sections and

immunohistochemistry studies (calretinin, cytokeratins5/6,

WT-1 antigen, polyclonal CEA, B72.3, Ber-Ep4).21 Addi-

tional eligibility criteria included: age B75 years, perfor-

mance status B2 according to the Eastern Cooperative

Oncology Group (ECOG), no significant comorbidities; no

extra-abdominal or hepatic metastases, and peritoneal dis-

ease amenable to potentially complete surgical cytore-

duction at preoperative computed tomography (CT) scan.22

The operative technique adopted in our centre was

detailed elsewhere.11 Briefly, the goal of the surgical cyto-

reduction was to remove all visible tumor by means of

parietal and pelvic peritonectomies with greater and lesser

omentectomy. Depending on disease involvement, multi-

visceral resections were performed, including cholecys-

tectomy, splenectomy, sigmoid colectomy, right/total

colectomy, and hysterectomy with salphingoophorectomy

in women.

Data on tumor distribution were prospectively collected

during laparotomy. Before cytoreduction, disease extent

was scored in 13 abdominopelvic regions according to the

Peritoneal Cancer Index (PCI), as outlined in Fig. 1.23 In

each region, a semiquantitative lesion size (LS) score was

used to rate the largest diameter of tumor deposits: LS-0:

no tumor; LS-1: B5 mm; LS-2: [5 mm and B50 mm; LS-

3: [50 mm. All the surgical specimens were submitted to

pathologic examination. Completeness of cytoreduction

(CC) was rated in each anatomical region according to

Sugarbaker et al. as complete cytoreduction (CC-0 = no

visible disease), near-complete cytoreduction (CC-

1 = residual disease B2.5 mm), or incomplete cytoreduc-

tion (CC-2 = residual disease [2.5 mm and B25 mm;

CC-3 = residual disease [25 mm).24 In the individual

patient, the results of the cytoreduction were rated

according to the worst CC score recorded in any region.

The HIPEC was performed according to the closed-

abdomen technique at a temperature of 42.5�C. Perfusate

volume was 4–6 L and average flow 700 mL/min. Drug

schedules were cisplatin (25 mg/m2/L) plus mitomycin-C

(3.3 mg/m2/L) for 60 min or cisplatin (43 mg/L) plus

doxorubicin (15.25 mg/L) for 90 min.25

Tumors were histologically categorized as epithelial,

biphasic or sarcomatoid, following the World Health

Organization (WHO) classification.26 Nuclear grade (NG)

was rated as follows: NG-1 = small nuclei, uniform

chromatin pattern, small nucleoli; NG-2 = larger nuclei,

some chromatin irregularity, more prominent nucleoli; NG-

3 = large nuclei, irregular chromatin pattern, prominent

nucleoli.27 Mitotic count per 50 high-power microscopic

fields was assessed.

All the patients underwent postoperative follow-up.

Clinical examination, thoracic/abdominal/pelvic CT scan,

and serum CA125 determination were performed every

3 months during the first 2 years and every 6 months

afterward. Additional studies were planned as necessary.

Study Design and Statistical Analysis

Both recurrence after complete or nearly complete cy-

toreduction and progressive tumor after suboptimal

cytoreduction were defined as disease progression. DMPM

recurrences were histologically documented by surgical

464 D. Baratti et al.

exploration (n = 12) or CT scan/ultrasound-guided biopsy

(n = 16). Alternatively, progressive DMPM was diagnosed

in comparison with the first examination after HIPEC

(n = 10), according to the Response Evaluation Criteria in

Solid Tumor Group (RECIST).28

The pattern of failure was analyzed in terms of presence

versus absence of tumor in 13 abdominal regions; extra-

abdominal involvement was also noted. Detailed informa-

tion regarding progressive disease distribution was

prospectively collected by CT scan in 38 patients. Pre-

liminary evaluation in 12 patients who underwent

laparotomy did not find statistical difference in disease

quantification by CT scan or surgical exploration (data not

shown). Only sites of disease involvement documented

within 3 months from first evidence of failure were inclu-

ded in the present analysis. This information forms the

basis of the present study.

Different spiral CT scanners were utilized over the 10-

year study period. However, a minimum 10-mm continu-

ous thickness was requested for all patients. All the studies

were performed with the administration of intravenous,

oral, and rectal contrast media. For the quantification of

disease extent, it was assumed that proximal and distal

jejunum were in the left-upper and left-lower quadrants,

while proximal and distal ileum were in the right-upper and

right-lower quadrants, respectively.29 For clinical use the

CT scans were read by staff radiologists; for the current

analysis, all CT scans were reread by the senior surgeon

(M.D.).

In 28 patients undergoing complete or nearly complete

cytoreduction, potential factors determining disease pro-

gression were statistically assessed for each anatomical

region. The following independent variables were ana-

lyzed: LS-score, CC-score (CC-0 versus CC-1), gender,

age at diagnosis, performance status at progression (ECOG

0/1 versus 2/3), histology (epithelial versus biphasic),

nuclear grade (NG-1/2 versus NG-3), mitotic count (B5

versus [5/50 HPF), surgery and systemic chemotherapies

prior to primary cytoreduction (done versus not done), time

to progression (TTP), ascites at primary cytoreduction

(present versus absent), and HIPEC drug schedule (cis-

platin and doxorubicin versus cisplatin and mytomicin-C).

Continuous variables were categorized into two classes

using their mean value as the cutoff. Fisher’s exact test and

Mann–Whitney U-test, as appropriate, were used for uni-

variate analysis and a logistic regression model for

multivariate analysis. Variables deemed statistically sig-

nificant by univariate analysis or considered clinically or

theoretically relevant for the purpose of the study,

regardless of the statistical significance, were included in

the model.

DMPM progression was treated according to the fol-

lowing general guidelines: repeated surgery was performed

in cases undergoing complete/nearly complete initial cy-

toreduction, with longer TTP ([4 months) and involving a

few localized lesions (B3) amenable to potentially com-

plete surgical resection; second HIPEC was performed

when multiple diffuse lesions were present; patients

FIG. 1 Peritoneal cancer index (PCI). The upper transverse line is

located at the costal margin and the lower at the anterior superior iliac

spine. Two sagittal lines divide the abdomen into three equal sectors.

Nine regions are defined. The small bowel is divided into four

additional regions. The anatomic structures involved in the 13

abdominopelvic regions are detailed. In each region, the greatest

diameter of peritoneal tumor implants is rated according to the

following semiquantitative score: lesion size (LS)-0 = no tumor; LS-

1 = B5 mm; LS-2 = [5 mm and B50 mm; LS-3 = [50 mm or

confluent smaller tumor nodules. The disease extent within all

regions is indicated by a numerical score from 0 to 39, obtained by

summing the LS score of each region

Recurrent Peritoneal Mesothelioma 465

lacking the above mentioned criteria underwent systemic

chemotherapy; patients with poor clinical conditions were

offered supportive care; emergency laparotomy was per-

formed as needed.

Survival from progression was dated from diagnosis of

postoperative disease progression to the time of death due

to any cause or last follow-up visit. Estimated survival

from progression was calculated by the Kaplan–Meier

method; the two-tailed log-rank test was used to assess the

significance of survival distributions.30 Based on univariate

analysis, significant variables were entered into a Cox

proportional hazard model for multivariate analysis.31 In

all statistical analyses P \ 0.05 was considered significant.

RESULTS

Clinical characteristics of the overall series are shown in

Table 1. Median TTP was 9 months [95% confidence

interval (CI) 1.6–35.9] for 38 patients (23 males and 15

females) with progressive DMPM after cytoreduction and

HIPEC. Disease progression occurred in 30 patients with

epithelial and 8 with biphasic DMPM; the patient with

sarcomatoid tumor is alive with no evidence of disease

after 12 months. There was a trend toward significance in

the difference of TTP between patients undergoing CC-0

(median 17 months) and CC-1 or CC-2/3 (median

8 months) initial surgical cytoreduction (P = 0.088). By

the time of the present analysis, 26 patients died, 10 were

alive with disease, and 2 were disease free. Median sur-

vival from progression was 8 months (95% CI 4–16.2)

(Fig. 3).

Pattern of Failure

In Fig. 2, mean LS- and CC-score at initial cytoreduc-

tion, along with progressive disease involvement following

combined treatment are displayed by anatomical region for

the overall series of 38 patients, according to the results of

the cytoreduction. Disease distribution at the time of pri-

mary cytoreduction was diffuse throughout the abdominal

cavity with relative sparing of small bowel and its mes-

entery for all the subgroups. By contrast, incomplete

surgical cytoreduction and progressive disease most fre-

quently involved the small bowel and its mesentery (in

patients categorized as CC-0, mean CC-score was 0 in all

anatomical regions).

In the individual patients, the pattern of failure was

categorized as follows: liver metastases occurred in one

patient, involvement of celiac and retroperitoneal lymph

nodes, respectively, in two, isolated seeding of the basal

pleura in two, and involvement of both abdominal and

pleural cavity in two. In the remaining 31 patients (81.6%)

only peritoneal progression was noted: the small bowel and

its mesentery were involved in 13 patients, intra-abdominal

sites exclusive of small bowel in 4, and both the small

bowel and additional intra-abdominal sites in 14. Overall,

small bowel was involved in 27 patients (71.1%). No

exclusive anastomotic suture line or laparotomy scar

involvement was recorded. Interestingly, no lymph-node

involvement was found at initial cytoreduction in the two

patients who developed nodal DMPM progression.

Factors Affecting the Pattern of Failure Following

Complete/Near-Complete Cytoreduction

Table 2 shows the results of the univariate and multi-

variate analysis of factors correlating to progressive disease

involvement in each anatomical area for the 28 patients

who underwent complete or near-complete cytoreduction.

At multivariate analysis, completeness of cytoreduction

correlated to disease progression in epigastric region [odds

ratio (OR) = 7.23; 95% CI = 1.05–49.69; P = 0.044],

upper jejunum (OR = 5.11; 95% CI = 1.12–23.28;

P = 0.035), lower jejunum (OR = 15.40; 95% CI = 2.5–

94.88; P = 0.003), and upper ileum (OR = 7.48; 95%

CI = 1.22–45.60; P = 0.029). Lesion size and previous

surgery were included in the model regardless of their

significance at univariate analysis. Lesion size correlated to

disease progression in upper jejunum, lower jejunum, and

upper ileum only at univariate analysis. Statistical signifi-

cance was not reached by any other clinicopathological

variable in any other region. The opening of hemidia-

phragms during surgery did not correlate to disease

progression in the respective pleural spaces (data not

shown).

Overall, disease progression occurred in 79 of 377

(20.9%) regions where macroscopically complete cytore-

duction was achieved, in 69 of 89 (77.5%) regions with

near-complete cytoreduction, and in 20 of 28 (71.4%) with

grossly incomplete cytoreduction. Distribution difference

was statistically significant (OR = 4.25; 95% CI = 2.93–

6.16; P \ 0.001).

Treatment and Outcome

Disease progression was treated by means of second

cytoreduction with HIPEC in three patients; of them, one is

presently disease free after 86 months from disease pro-

gression, one died after 30 months, and one is alive with

disease after 17 months. Four patients underwent surgical

debulking; one is presently disease-free after 30 months,

two are alive with disease after 2 and 30 months, respec-

tively, and one died after 4 months. Sixteen patients

underwent systemic chemotherapy. Drug schedule was

cisplatin and pemetrexed in seven patients, cisplatin and

gemcitabine in three, cisplatin and doxorubicin in one,

466 D. Baratti et al.

cisplatin alone in one, and vinorelbine in four. No major

response was observed. By the time of the present analysis,

ten patients died and six are alive. Median survival from

progression was 11 months (range 6–32 months) in this

group. Only supportive cares were offered to 15 patients:

14 of them died after a median of 2 months (range

0–26 months) and only one is currently alive.

Results from univariate and multivariate analysis of

factors influencing survival from progression (SFP) are

shown in Table 3. Poor performance status, progression-

free interval B9 months, and supportive versus curative

treatment (second cytoreduction with HIPEC, surgical

debulking or systemic chemotherapy) were independent

predictors of reduced SFP.

DISCUSSION

To our knowledge, this is the first paper to extensively

address the issue of treatment failure in patients with

DMPM undergoing cytoreduction and HIPEC. The present

analysis provided some interesting observations on the

pattern of failure: first, DMPM progression after compre-

hensive treatment most often remained confined to the

abdominal cavity; second, although tumor distribution

before initial cytoreduction was diffuse throughout the

abdomen and pelvis with relative sparing of the small

bowel, both residual and progressive disease after cytore-

ductive surgery most commonly involved the small bowel

with its mesentery. This is consistent with the notion that

local-regional chemotherapy is largely ineffective against

gross peritoneal disease, presumably due to its limited

tumor penetration.32 Accordingly, incomplete cytoreduc-

tion results in disease progression in the area of

macroscopic residual tumor. These predictable surgical

failures could be prevented only by careful patient

selection.

In the present analysis, patients who were able to

undergo optimal cytoreduction were analyzed separately

and residual disease B2.5 mm (as compared with no visi-

ble tumor) was the only independent risk factor for

TABLE 1 Clinical characteristics of 70 patients with diffuse malignant peritoneal mesothelioma according to the results of initial cytoreduction

Variables Categories CC-0 CC-1 CC-2/3 Overall series

No. of patients 26 30 14 70

Disease progression 11 17 10 38

Operative death/other cause 1 – 2 3

Gender Male 10 16 6 32

Female 16 14 8 38

Median age at diagnosis,

years (range)

40 (24–74) 54 (22–76) 56 (40–66) 52 (22–76)

Histology Epithelial 23 25 13 61

Biphasic 3 4 1 8

Sarcomatoid – 1 – 1

Median interval diagnosis/HIPEC,

months (range)

5 (1–81) 3 (1–23) 5 (1–11) 4 (1–81)

Surgical procedures before

cytoreduction and HIPEC

Only biopsy 11 19 10 40

1 region dissected 11 7 1 19

2–5 regions dissected 4 4 3 11

[5 regions dissected – – – –

Systemic CT before

cytoreduction and HIPEC

Done 18 20 11 49

Not done 8 10 3 21

PCI, mean (range) 15.0 (3–30) 23.3 (3–39) 33.8 (30–39) 21.4 (3–39)

HIPEC schedule Cisplatin ? doxorubicin 23 27 9 59

cisplatin ? mitomycin-C 3 3 5 11

Median time-to-progression,

months (range)

17 (2–57) 8 (2–81) 8 (2–28) 9 (2–81)

Performance at progression ECOG 0/1 9 15 9 32

ECOG 2/3 2 2 1 6

CC-0 = macroscopically complete surgical cytoreduction; CC-1 = nearly complete surgical cytoreduction (residual disease B2.5 mm);

CC-2/3 = grossly incomplete cytoreduction (residual disease [25 mm); PCI, peritoneal cancer index; CT, chemotherapy; HIPEC, hyperthermic

intraperitoneal chemotherapy; ECOG, Eastern Cooperative Oncology Group

Recurrent Peritoneal Mesothelioma 467

progressive disease involvement of the epigastric region,

upper ileum, upper and lower jejunum. These findings

provide interesting information regarding the significance

of the completeness of cytoreduction in the comprehensive

treatment of DMPM, suggesting that failure to remove all

the visible tumor is related to disease progression. The

relative contribution of HIPEC to cytoreductive surgery is

still poorly known, Only indirect pieces of evidence come

FIG. 2 Mean lesion size (a) and

completeness of cytoreduction (b)

scores and disease involvement at

progression after cytoreduction

and HIPEC (c) by anatomical

region according to the results of

the initial cytoreduction. The

completeness of the cytoreduction

(CC) was indicated as CC-0 = no

visible disease; CC-1 = residual

disease B2.5 mm;

CC-2 = residual

disease [2.5 mm and B25 mm;

CC-3 = residual

disease [25 mm. Progressive

disease involvement was

expressed as the percentage

of patients with documented

progressive disease in each

abdominal region

468 D. Baratti et al.

from experimental data.32 Furthermore, in clinical studies

which reviewed potential prognostic indicators for sur-

vival, the results of the cytoreduction (determined in each

single patient according to the maximum thickness of

residual tumor in any place of the abdomen) have been

repeatedly demonstrated to be closely associated to sur-

vival.7–12 In this series, an innovative study approach with

a more detailed analysis of the distribution of progressive

disease was performed and potential variables correlating

to the study end-point (i.e., disease progression) were

assessed at the level of the single anatomical regions, not of

the individual patients.

The epigastric region and the small bowel are critical

areas where adequate cytoreductive surgery is technically

difficult.16–20 In these regions, lesion size correlated to

disease progression at univariate but not at multivariate

analysis. This suggests that disease amount and com-

pleteness of cytoreduction might be related in this patient

population. Such hypothesis is consistent with the obser-

vations of Yan et al.29 The authors demonstrated that

massive DMPM involvement of the epigastric region and

the small bowel exclusive of the distal ileum at preoper-

ative CT scan predicted incomplete cytoreduction. The

peritoneal anatomy of the epigastric/subhepatic area is

complex. Large tumor involving the hepatic hilum ham-

pers tumor removal. Disease dissection from subpyloric

space and the lesser omental vascular arcade may com-

promise the blood supply to the stomach, unless total

gastrectomy is performed. Analogously, distal ileum can

be easily resected and ileocolic anastomosis performed,

but only limited small-bowel resections or local electro-

surgical tumor dissection can be performed in proximal

segments, resulting in persisting disease and high rates of

postoperative failure.

In the literature, the definition of adequate cytoreduction

for peritoneal mesothelioma is controversial. In our and

Lyon group series, cytoreduction down to persisting dis-

ease B2.5 mm correlated to better survival, compared with

1.0

SurvivalProbability

0.8

0.6

0.4

0.2

0 2412 18Months from Disease Progression

6

Patientsat Risk

38 915 1025

FIG. 3 Survival from progression in 38 patients with diffuse

malignant peritoneal mesothelioma

TABLE 2 Univariate and multivariate analysis of factors affecting progressive disease involvement in 28 patients undergoing initial complete/

nearly complete cytoreduction by anatomical region

Completeness of cytoreduction

CC-1 versus CC-0

Lesion size

LS-3 versus LS-2 versus LS-1 versus LS-0

Univariate Multivariate Univariate Multivariate

P value OR (95% CI) P value P value P value

0. Central NA 1.77 (0.65–11.75) 0.643 NS NS

1. Right upper 0.443 1.87 (0.59– 5.89) 0.280 NS NS

2. Epigastric 0.001 7.23 (1.05–49.69) 0.044 NS NS

3. Left upper 0.527 1.57 (0.61–4.15) 0.355 NS NS

4. Left flank 1.00 1.68 (0.55–5.06) 0.356 NS NS

5. Left lower 1.00 1.54 (0.56–4.24) 0.396 NS NS

6. Pelvis NA 1.19 (0.47–2.99) 0.708 NS NS

7. Right lower 0.339 1.03 (0.51–2.07) 0.933 NS NS

8. Right flank 1.00 0.85 (0.27–2.72) 0.796 NS NS

9. Upper jejunum 0.001 5.11 (1.12–23.28) 0.035 0.043 NS

10. Lower jejunum 0.001 15.40 (2.5–94.88) 0.003 0.037 NS

11. Upper ileum 0.008 7.48 (1.22–45.60) 0.029 0.051 NS

12. Lower ileum 0.210 3.82 (0.871–16.81) 0.075 0.099 NS

Completeness of cytoreduction: CC-1 = residual disease B2.5 mm; CC-0 = no visible residual tumor; lesion size: LS-3 [50 mm; LS-2 [5 mm

and B50 mm; LS-1 B5 mm; LS-0 = no tumor; NA not applicable; NS not significant; OR odds ratio; CI confidence interval

Recurrent Peritoneal Mesothelioma 469

larger residual tumor.10,11 In the series of the Bethesda and

the Washington Cancer Center, increased survival corre-

lated to residual disease B10 mm and B25 mm,

respectively.8,9 Active doxorubicin transportation into

mesothelioma cells or their better chemosensitivity and

permeability might explain these findings.33 However, a

broader definition of adequate cytoreduction would imply

that less aggressive cytoreductive surgery, leaving behind

relatively larger residual tumor, may be appropriate.

Apparently, our findings suggest opposite conclusions and

support the absolute requirement for maximal surgical

efforts to remove all visible tumor.

The 2.5-mm cutoff between optimal and incomplete

cytoreduction is based on the maximal tumor penetration of

intraperitoneal cisplatin.32 This value was obtained in mice

under experimental conditions, being not necessarily true

in the operating room. Different drugs, temperatures, car-

rier solutions or perfusion techniques may imply different

tumor penetration. Pharmacokinetic studies to determine in

vivo drug penetration during closed-abdomen HIPEC are

presently ongoing in our center.

Our findings may have practical implications on the

surgical strategy for DMPM. Pseudomyxoma peritonei is

another peritoneal surface malignancy successfully man-

aged by cytoreduction and HIPEC. Patients with extensive

perigastric involvement undergo total gastrectomy and

their survival results are comparable to those of any other

with complete cytoreduction.34 Conversely, total gastrec-

tomy is not currently performed for DMPM and extensive

perigastric disease is associated to inadequate cytoreduc-

tion and poor prognosis. Since the comprehensive

management of DMPM seems to improve with subsequent

reports, our data suggest that total gastrectomy may be

considered for these patients.

In the present study, progressive disease distribution was

assessed by CT scan, being confirmed by surgical explo-

ration only in a minority of patients. This may imply a

limitation in the quality of our data. However, CT scan has

been shown to reliably identify size and anatomical dis-

tribution of DMPM in the selection process for combined

treatment.29 Furthermore, a recent series of 28 patients with

various peritoneal surface malignancies demonstrated that

CT scan measurement of peritoneal cancer index is accu-

rate as intraoperative direct assessment.35

Poor performance status and short disease-free interval

from initial cytoreduction were independently related to

decreased survival from progression. Patients with early

progressive tumor and compromised clinical conditions are

thought to have aggressive disease and they are likely

excluded from challenging treatment. These data suggest

that outcome following failure of primary combined

treatment is primarily determined by inherent biologic

tumor features. Nevertheless, biphasic histology correlated

to reduced survival after progression at univariate but not

TABLE 3 Univariate and multivariate analysis of factors affecting survival from progression in 38 patients

Variables Univariate Multivariate

P value Odds ratio (95% CI) P value

Sex (male versus female) 0.807

Age at diagnosis (B52 versus [52 years) 0.841

ECOG at progression (0/1 versus 2/3) \0.001 3.33 (2.57–4.97) 0.005

Surgical procedures before cytoreduction and HIPEC (B1 versus [1 reg. dissected) 0.736

Histology (epithelial versus biphasic) 0.045 1.14 (0.40–3.24) 0.801

PCI at initial cytoreduction (B20 versus [20) 0.390

Ascites at initial cytoreduction (present versus absent) 0.910

CC at initial cytoreduction (0/1 versus 2/3) 0.532

Pattern of failure (intra- versus extra-abdominal) 0.351

Therapy (supportive versus curativea) \0.001 5.26 (1.72–14.28) 0.003

Nuclear grade (1/2 versus 3) 0.142

Mitotic count (B5 versus [5/50 HPF) 0.142

Time to progression ([9 versus B9 months) \0.001 3.84 (1.38–11.11) 0.009

Systemic CT before cytoreduction and HIPEC (done versus not done) 0.786

Interval diagnosis/HIPEC (B6 versus [6 months) 0.874

CI 95% confidence interval; ECOG Eastern Cooperative Oncology Group performance score; HIPEC hyperthermic intraperitoneal chemo-

therapy; PCI peritoneal cancer index; CC completeness of cytoreduction score; CT chemotherapya Second cytoreduction and HIPEC, debulking surgery, systemic chemotherapy

470 D. Baratti et al.

multivariate analysis, probably owing to the imbalance of

the proportion of cases with epithelial or biphasic DMPM.

Curative versus supportive second-line treatment cor-

related to increased survival from progression at

multivariate analysis. Furthermore, operative treatment

(i.e., cytoreduction or cytoreduction with HIPEC) showed a

trend toward better outcome, compared with systemic

chemotherapy. Since treatment was determined according

to patient conditions and disease extent, a selection bias

could have likely occurred in this setting and the different

treatment modalities cannot be compared. Nevertheless,

encouraging results were obtained with repeated cytore-

duction and HIPEC. Along with the tendency of DMPM to

remain confined to the peritoneal surface throughout its

clinical course, these data make aggressive local-regional

treatment of progressive disease an attractive option.

In conclusion, failure to surgically remove all the visible

disease in critical anatomical regions was the leading cause

of DMPM progression after combined treatment. Accord-

ingly, clinical results may improve with better patient

selection, increased surgical aggressiveness, and the future

development of more effective local-regional, systemic or

targeted therapy. In selected patients, management of

progressive DMPM by means of reoperative surgery,

possibly followed by repeated HIPEC, seems worthwhile.

ACKNOWLEDGEMENTS This study was supported in part by

grants from the Italian Association for Cancer Research (AIRC) and

the Italian Health Ministry.

REFERENCES

1. Hassan R, Alexander R, Antman K, et al. Current treatment

options and biology of peritoneal mesothelioma: meeting sum-

mary of the first NIH peritoneal mesothelioma conference. AnnOncol. 2006;17:1615–9.

2. Boffetta P. Epidemiology of peritoneal mesothelioma: a review.

Ann Oncol. 2007;18:985–90.

3. Robinson BW, Lake RA. Advances in malignant mesothelioma.

N Engl J Med. 2005;353:1591–603.

4. Zaffaroni N, Costa A, Pennati M, et al. Survivin is highly

expressed and promotes cell survival in malignant peritoneal

mesothelioma. Cell Oncol. 2007;29:453–66.

5. Borczuk AC, Cappellini GCA, Kim HK, et al. Molecular profil-

ing of malignant peritoneal mesothelioma identifies the ubiquitin-

proteasome pathway as a therapeutic target in poor prognosis

tumor. Oncogene. 2007;26:610–7.

6. Janne PA, Wozniak AJ, Belani CP, et al. Open-label study of

pemetrexed alone or in combination with cisplatin for the treat-

ment of patients with peritoneal mesothelioma: outcomes of an

expanded access program. Clin Lung Cancer. 2005;7:40–6.

7. Loggie BW, Fleming RA, McQuellon RP, et al. Prospective trial

for the treatment of malignant peritoneal mesothelioma, Am Surg.

2001;67:999–1003.

8. Sugarbaker PH, Welch LS, Mohamed F, et al. A review of per-

itoneal mesothelioma at the Washington Cancer Institute. SurgOncol Clin North Am. 2003;12:605–21.

9. Feldman AL, Libutti SK, Pingpank JF, et al. Analysis of factors

associated with outcome in patients with malignant peritoneal

mesothelioma undergoing surgical debulking and intraperitoneal

chemotherapy. J Clin Oncol. 2003;21:4560–7.

10. Brigand C, Monneuse O, Mohamed F, et al. Malignant peritoneal

mesothelioma treated by cytoreductive surgery and intraperito-

neal chemohyperthermia: results of a prospective study. Ann SurgOncol. 2006;13:405–12.

11. Deraco M, Nonaka D, Baratti D, et al. Prognostic analysis of

clinicopathologic factors in 49 patients with diffuse malignant

peritoneal mesothelioma treated with cytoreductive surgery and

intraperitoneal hyperthermic perfusion. Ann Surg Oncol.2006;13:229–37.

12. Yan TD, Brun EA, Cerruto CA, et al. Prognostic indicators for

patients undergoing cytoreductive surgery and perioperative

intraperitoneal chemotherapy for diffuse malignant peritoneal

mesothelioma. Ann Surg Oncol. 2007;14:41–9.

13. Markman M, Kelsen D. Efficacy of cisplatin-based intraperito-

neal chemotherapy as treatment of malignant peritoneal

mesothelioma. J Cancer Res Clin Oncol. 1992;118:547–50.

14. Neumann V, Muller KM, Fischer M. Peritoneal mesothelioma—

incidence and aetiology. Pathologe. 1999;20:169–76.

15. Eltabbakh GH, Piver MS, Hempling RE, et al. Clinical picture,

response to therapy, and survival of women with diffuse malig-

nant peritoneal mesothelioma. J Surg Oncol. 1999;70:6–12.

16. Zoetmulder FAN, Sugarbaker PH. Patterns of failure following

treatment of pseudomyxoma peritonei of appendiceal origin. EurJ Cancer. 1996;32:1727–33.

17. Verwaal VJ, Boot H, Aleman BM, et al. Recurrences after peri-

toneal carcinomatosis of colorectal origin treated by

cytoreduction and hyperthermic intraperitoneal chemotherapy:

location, treatment, and outcome. Ann Surg Oncol. 2004;11:

375–9.

18. Smeenk RM, Verwaal VJ, Antonini N, et al. Progression of

pseudomyxoma peritonei after combined modality treatment:

management and outcome. Ann Surg Oncol. 2007;14:493–9.

19. Bjelic L, Yan TD, Sugarbaker PH. Failure analysis of recurrent

disease following complete cytoreduction and perioperative

intraperitoneal chemotherapy in patients with peritoneal carci-

nomatosis from colorectal cancer. Ann Surg Oncol. May

15;[Epub ahead of print].

20. Yan TD, Bjelic L, Sugarbaker PH. Critical analysis of treatment

failure after complete cytoreductive surgery and perioperative

intraperitoneal chemotherapy for peritoneal dissemination from

appendiceal mucinous neoplasms. Ann Surg Oncol. 2007 June

1;[Epub ahead of print].

21. Nonaka D, Kusamura S, Baratti D, et al. Diffuse malignant

mesothelioma of the peritoneum. Cancer. 2005;104:2181–8.

22. Oken MM, Creech RH, Tormey DC, et al. Toxicity and response

criteria Of The Eastern Cooperative Oncology Group. Am J ClinOncol. 1982;5:649–55.

23. Esquivel JE, Sugarbaker PH. Elective surgery in recurrent colon

cancer with peritoneal seeding: when to and when not to. CancerTher. 1998;1:321–5.

24. Jaquet P, Sugarbaker PH. Current methodologies for clinical

assessment of patients with peritoneal carcinomatosis. J Exp ClinCancer Res. 1996;15:49–58.

25. Rossi CR, Foletto M, Mocellin S, et al. Hyperthermic intraop-

erative intraperitoneal chemotherapy with cisplatin and

doxorubicin in patients who undergo cytoreductive surgery for

peritoneal carcinomatosis and sarcomatosis: phase I study. Can-cer. 2002;94:492–9.

26. Battifora H, McCaughey WTE. Tumors of the Serosal Mem-

branes. Washington, DC: Armed Forces Institute of Pathology;

1994. Atlas of Tumor Pathology; 3rd series, fascicle 15.

27. Goldblum J, Hart WR. Localized and diffuse mesotheliomas of

the genital tract and peritoneum in women. A clinicopathologic

study of nineteen true mesothelial neoplasms, other than

Recurrent Peritoneal Mesothelioma 471

adenomatoid tumors, multicystic mesotheliomas, and localized

fibrous tumors. Am J Surg Pathol. 1995;19:1124–37.

28. Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to

evaluate the response to treatment in solid tumors. European

Organization for Research and Treatment of Cancer, National

Cancer Institute of the United States, National Cancer Institute of

Canada. J Natl Cancer Inst. 2000;92:205–16.

29. Yan TD, Haveric N, Carmignani CP, et al. Abdominal computed

tomography scans in the selection of patients with malignant

peritoneal mesothelioma for comprehensive treatment with

cytoreductive surgery and perioperative intraperitoneal chemo-

therapy. Cancer. 2005;103:839–49.

30. Kaplan EL, Meier P. Nonparametric estimation from incomplete

observations. J Am Stat Soc. 1958;53:457–81.

31. Cox DR. Regression models and life tables (with discussion). J RStat Soc B. 1972;34:187–220.

32. Los G, Mutsaers PH, van der Vijgh WJ, et al. Direct diffusion of

cis-platinum in intraperitoneal rat tumors after intraperitoneal

chemotherapy: a comparison with systemic chemotherapy. Can-cer Res. 1989;48:3380–4.

33. Yan T, Welch L, Black D, et al. A systematic review on the

efficacy of cytoreductive surgery combined with perioperative

intraperitoneal chemotherapy for diffuse malignancy peritoneal

mesothelioma. Ann Oncol. 2007;18:827–34.

34. Sugarbaker PH. Cytoreduction including total gastrectomy for

pseudomyxoma peritonei. Br J Surg. 2002;89:208–12.

35. Gann M, Pitroff A, Esquivel J. prospective comparison of the

peritoneal cancer index at the time of surgery in 28 patients

with peritoneal surface malignancies. Ann Surg Oncol. 2007;

14(supl):P75.

472 D. Baratti et al.