2. OBJECTIVES
2.1. Primary endpoint
The present study will compare LRT followed by second line chemotherapy with only second line systemic chemotherapy in patients affected by stage III/IV epithelial ovarian cancer with macroscopic residual disease after 1st line chemotherapy, in terms of overall survival.
2.2. The secondary endpoints
a To evaluate the progression free survival calculated form the date of randomisation;
b To analyse LRT related morbidity, mortality and toxicity with identification of risk factors associated with development of operative complications among the following variables: performance status, age, body surface, extension of secondary CRS, performance of bowel anastomosis, duration of procedure, carcinomatosis extension, residual disease after secondary CRS, occurrence of toxicity during the first line chemotherapy;
c To evaluate the extent to which the performance of LRT increases toxicities rates of subsequent second line chemotherapy. This will be accomplished through the comparative evaluation of toxicity between the 2 arms during the second line therapy;
d To identify prognostic and predictive factors for response to therapy in the experimental arm among the following clinicopathologic variables: age of patients, performance status, residual disease after the primary surgery, disease extension at SLO, residual disease after secondary cytoreduction at SLO, tumour grade and histological subtype.
e Moreover, in consideration of emerging role of biomolecular markers in predicting clinical outcome following specific treatments, the following biological variables – which proved to be involved in determining the cellular response to platinum analogues and/or taxanes - will be evaluated on tumour specimens obtained during CRS:
? p53 accumulation
? apoptosis-related markers (bcl-2, bcl-xl, bax, survivin)
? drug-resistance associated markers (MDR1, MRP)
? detoxification enzymes (GST-?)
? ERCC1 as an indicator of nucleotide excision repair.
The expression of ERCC1 mRNA will be evaluated on frozen specimens, whereas for the other investigated biomarkers the evaluation will be carried out at a protein level on formalin-fixed, paraffin-embedded specimens using commercially available reagents. Whenever feasible, according to the availability of residual tumor material, the expression of these latter markers will be determined also at a molecular level.
In addition to the evaluation of molecular determinants of chemoresistance, the determination of DNA ploidy by flow cytometry (on nuclei obtained from formalin-fixed, paraffin-embedded sections or from frozen specimens) will be also carried out. In fact, DNA ploidy represents the only biological variable whose determination has been included in the clinical recommendation of the 1998 Consensus Statement, for its association with ovarian cancer aggressiveness (Annals of Oncology, 10, 1999).
Whenever feasible, all these determinations will be repeated

3. BRIEF DESCRIPTION OF STUDY
3.1 Study design
The current investigation is a prospective multicentric randomised clinical trial that will evaluate the impact in terms of survival of secondary CRS followed by IPHP in the treatment of macroscopically positive second look stage III/IV epithelial ovarian cancer (see Appendix, Figure 1).
Patients with pathologically verified stage III/IV epithelial invasive ovarian cancer will receive 6 cycles of cisplatin based chemotherapy every 3 weeks (see section 6.1 for more details). Four weeks after the end of treatment, eligible patients, with evidence of partial or no response, but optimally resectable disease evaluated by abdominopelvic CT scan according to criteria described by Nelson et al [46] (see Appendix, table 10) will be randomly allocated in one of the following two treatment arms: 1) LRT arm (secondary cytoreduction associated with IPHP) followed by second line systemic therapy; 2) control arm which consists of second line systemic therapy (see Appendix, Figure 1).
3.2 Variables definition
Variables related to the patient
3.1. Age: age of the patient at the time of randomisation counted in years;
3.2. Performance status: defined according to the ECGO criteria (see appendix, Table 9);
4. Weight: measured (kg) at the time of procedure;
5. Height: measured (cm) at the time of procedure;
6. Body surface area (BSA): It will be calculated by the following equation:
BSA (m2): 0,024265 x height 0,3964 x weight 0,5378
Variables related to the tumour
6.1. Histological subtype; defined according to WHO criteria;
6.2. Tumour grade: well, moderately, or poorly differentiated, according to pattern grading classification [166,167];
6.3. Disease staging: will be carried out according to the FIGO criteria;
6.4. Malignant ascites: presence of neoplastic cells in the peritoneal liquid or washing confirmed by cytological evaluation;
Variables related to the primary therapy
6.5. Residual disease (RD) at primary surgery: < 1 cm and > 1 cm;
6.6. Toxicity rate during the first line chemotherapy: percentage of patients developing grade I, II, III and IV toxicities, during the adjuvant chemotherapy, defined according to the WHO criteria (see Appendix, Table 6). Only the toxicities grades III/IV will be considered as dependent variable for statistical analysis;
Variables related to the LRT
6.7. Extension of carcinomatosis: defined according to Sugarker’s criteria for Peritoneal cancer index [168] (see appendix, figure 3);
6.8. Number of peritonectomy procedures: can vary between 1 to 6 procedures (see section 6.2.1);
6.9. Extension of procedure: defined according to the number of peritonectomy procedures in three categories: level -1 (less than 2 peritonectomy procedures), level -2 (two to four peritonectomy procedures) and level -3 (Complete Peritonectomy);
6.10. Duration of procedure: calculated in minutes from the moment of anaesthetic induction until the abdominal wall closure;
6.11. Residual disease (RD) after secondary cytoreduction: defined according to Sugarbaker’s criteria. in 4 classes [168]: CC-0 - no residual disease, CC-1 - minimal residual disease <2.5 mm, CC-2: residual disease 2.5mm-2.5 cm, CC-3: residual disease >2.5 cm. (see appendix, figure 3); Optimal cytoreduction is defined as RD < 2.5mm (cc-0 and cc-1); sub optimal cytoreduction as RD > 2.5 mm;
6.12. Number of bowel anastomosis: counted during the secondary cytoreduction;
Variables related to complications associated with LRT
Only the events occurring until the 28th day in the postoperative period will be considered.
6.13. Toxicity rate: percentage of patients developing grade I, II, III and IV toxicities defined according to the WHO criteria (see Appendix, Table 6);
6.14. Morbidity rate (see appendix, Table 8): percentage of patients developing grade I, II, III and IV morbidity; Only the morbidities grades III/IV will be considered as dependent variable for statistical analysis;
6.15. Mortality rate: percentage of patients evolving to death due to complication related to the LRT;
Variables related to follow-up and response to therapy
6.16. Response to therapy will be classified in one of the following categories (see appendix, Table 7):
6.16.1.1.1. complete response;
6.16.1.1.2. partial response;
6.16.1.1.3. stable disease;
6.16.1.1.4. disease progression. This evaluation will be applied both during the first and second line chemotherapies;
6.17. Follow-up length: period of time measured in months from the date of randomisation and occurrence date of death or disease relapse or last contact whichever occurs first;
6.18. Unfavorable events:
a) Death of disease: death of patient due to disease progression;
b) Intercurrent death: death due to cause other than ovarian cancer;
c) Recurrence of disease: relapse of disease detected during the follow-up by clinical examination and/or CT scan and/or Ca 125 and/or chest radiography. Can be loco regional or distant;
d) Disease progression: patient will be classified as having disease progression in case of disease recurrence or death, whichever occurs first.
4. ELIGIBILITY CRITERIA
The following criteria for inclusion or exclusion of patients from the study will be checked before the randomisation.
1. histological diagnosis of stage III/IV epithelial invasive ovarian carcinoma or carcinoma of fallopian tube;
2. age less than 75 years;
3. patients who were submitted to primary surgical staging and front line primary chemotherapy;
4. partial response to first line chemotherapy with persistent disease during the adjuvant treatment;
5. recurrences within 6 months from the end of first line chemotherapy;
6. leukocyte count >3500/mm3, neutrophyl count >1500/mm3, and platelet count > 100.000/mm3;
7. adequate renal function with serum creatinine level < 1.5 mg/dl;
8. performance status (ECOG) 0, 1 or 2;
9. informed consent from the patients;
10. cytoreducible disease evaluated by abdominopelvic CT scan [46];
Patients will be excluded from the study when:
1. Epithelial ovarian cancer treated by more than one line chemotherapy regimen;
2. Epithelial borderline ovarian tumours;
3. Impossibility of adequate follow up;
4. Other malignancy except adequately treated basal cell skin cancer or carcinoma in situ of the cervix;
5. Active infection or other serious underlying conditions that would impair the ability of patient to receive the protocol treatment;
6. Relapse after 6 months from the end of first line chemotherapy;
7. Liver, pulmonary or brain metastasis;
8. Pleural naoplastic effusion;
9. Inadequate renal function with serum creatinine > 1.5mg/dl;
10. Serum transaminases > 2.5 times the upper level of normal;
11. Serum bilirrubin > 2 mg/dl;
12. Inadequate medullar haematopoiesis;
13 Psychological, familial, sociological, or geographical condition that would preclude study;
14 Complete bowel obstruction.
5. TREATMENT ASSIGNMENT
The random treatment allocation constitutes a crucial component of a controlled clinical trial in so far as it minimizes bias and guarantees the validity of probabilistic method that underlies the inferential process. Randomised trials, if done badly are open to bias and thus, we followed the guidelines established by CONSORT statement [169] to report the randomisation process.
Randomisation will be done by means of computer generated blocked randomisation list stratified according to participating centers.
The concealment of generated randomised sequence will be guaranteed by the adoption of a centrally administered telephone-based assignment scheme [170,171]. The randomisation process will be executed by a central office located at a Randomisation and data processing committee (RDPC) (see section 10). The allocation process will require the participating centre to initiate the request. This will be done the day before the second-look operation, by sending the form, completed for a patient. It will contain the following information:
a) Name, ID number, age, data / local of birth and address of the patient;
b) Checks for eligibility;
c) Check to determine if the patient had signed the study informed consent and had indicated his willingness to accept either surgical or medical treatment;
d) Check the willingness of medical staff responsible for the treatment to accept the randomisation.
If the patient is deemed as candidate for randomisation after fulfilling the previous items, the Randomisation and Data processing committee (RDPC) will be contacted by telephone to define the treatment allocation.
An allocation will not be released by the RDPC if essentials items of information are missing from the forms, if an eligibility stop condition is checked, or if the clinic does not indicate that the signed informed consent is not obtained from the patient. The person responsible for the patient’s information checks, just before the randomisation, must be different from the one who generates the random sequence [170,171].
Each participating centre wil have to send written confirmation of patient entry and treatment assignment to the RDPC, after the telephone randomisation.
We considered masking not feasible in this study as it is a surgical trial. Exclusions from the study after randomisation will be avoided but if it eventually occurs it will be clearly documented including their reasons (eventual discovery of ineligibility of a participant, deviations from the protocol, withdrawals) [172]. All randomised participants in the originally assigned groups, regardless of compliance to protocol will be analysed following the intention to treat policy (see section 9.2).
6. TREATMENT DESCRIPTION
6.1. Primary chemotherapy
After the primary surgical staging, patients with stage III/IV invasive ovarian cancer will receive adjuvant cisplatin based chemotherapy. An advisable combination is carboplatin AUC 5-7 associated with paclitaxel 175 mg/m2 with a 3-hour infusion (every 3 weeks for 6 cycles). At this point the patient will be considered potential candidate for protocol enrolment.
The patients will have to have a white-cell count of at least 3,000/mm3 and a platelet count of at least 100,000/mm3 before the next course could be administered. Courses will be delayed week by week until these counts are achieved. If this delay exceeds three weeks patient will be withdrawn from the study. After the 6th cycle, a clinical evaluation will be performed (general and gynaecologic physical examination, abdominopelvic CT scan and sonography, chest W-ray and serum Ca 125) for assessment of tumour response to therapy.
6.2. LRT
If the patient is classified as having an optimally cytoreducible disease, by preoperative abdominopelvic CT scan [46], she will be randomly assigned to one of treatments arms: the LRT or control arms (see section 5). Otherwise the patient will be referred for other study protocols. The reason for the exclusion lies in the fact that the maximization of IPHP effectiveness can be attained in patients with minimal residual disease, as it was previously discussed (see section 1.6). Furthermore, an unresectable carcinomatosis even after 6 cycles of adjuvant chemotherapy must be classified as refractory and progressive disease, and as it was presented before, a secondary cytoreduction under this condition is not advisable [75,76].
The LRT arm will comprise the secondary CRS followed by IPHP and second line chemotherapy. The control arm will comprise the second line therapy only.
6.2.1 Secondary CRS
The peritonectomy procedure described by Sugarbaker will be adopted [103].
6.2.1.1 Preparation
At the time in which patient is allocated to the LRT arm she must have been submitted to:
• general and gynaecological examination;
• Imaging assessment: abdominopelvic CT scan and abdominopelvic sonography, Chest X ray;
• laboratory exams: serum Ca-125, complete blood cell count, serum albumin, creatinine clearance.
Two days before the procedure the patient will start the colon clearance. In the day before surgery she will have to have a central venous access.
In the operating room, the patient will be put in a supine position with gluteal folds advanced to the break on the operating table to allow full access to the perineum during the surgical procedure. This position is essential to avoid intraoperative skin or muscle necrosis. The weight of the legs must be directed to bottom of the feet by positioning the footrests so that minimal weight is borne by the calf muscle. Myonecrosis within the posterior compartment of the leg may occur unless the legs are protected properly. A 3-way bladder catheter and a large-bore silastic nasogastric tube are positioned.
Abdominal skin preparation will be from mid chest to mid thigh as well as the external genitalia, including vagina. The abdomen will be opened from xyphoid to pubis. Generous abdominal exposure will be achieved through the use of a Thompson Self-Retaining Retractor.
6.2.1.2 Laser-mode electro surgery
A ball-tip electrosurgical handpiece will be used to dissect the tumour on peritoneal surfaces from normal tissue [173]. The electrosurgery will be used on pure cut at high voltage. The 2 mm ball-tip electrode is used for dissecting on visceral surfaces, including stomach, small bowel, and colon. When more rapid tumour destruction is required, the 5 mm ball-tip can be used.
6.2.1.3 Surgical Steps
Each procedure that composes the peritonectomy technique has a definite resection that requires an orderly sequence of surgical maneuvers to create an optimum cytoreduction. One or more of following steps can be performed depending on the extension of primary surgical staging or disease extension at the time of SLO, in order to achieve optimal residual status.
• greater omentectomy, right parietal peritonectomy and right colon resection
• left upper quadrant peritonectomy, splenectomy and left parietal peritonectomy
• right upper quadrant peritonectomy and Glissonian’s capsule resection
• lesser omentectomy, colecystectomy, stripping of omental bursa and antrectomy
• pelvic peritonectomy with sigmoid colon resection with or without hysterectomy and bilateral salpingo-oophorectomy;
• other intestinal resection and/or abdominal mass resection.
• bowel anastomosis (this step can also be performed after the completion of IPHP, see section 1.5.3.1).
6.2.2 Intraperitoneal hyperthermic perfusion (IPHP)
After secondary CRS, 4 Tenckhoff catheters will be placed in the abdominal cavity. Two inflow catheters will be placed in the right subphrenic cavity and at deep pelvic level, respectively; and two further catheters in the left subphrenic cavity and in the superficial pelvic site.
6.2.2.1 The device
The IPHP requires the employment of lung-heart machine, comprised by a roller pump, a thermostat, a heat exchanger and an extra corporeal circuit. The perfusate flow will be controlled as well as the heat exchanger adjusts the temperature of perfusate, by circulating water at a desired temperature in the arterial phase of circuit. The extra corporeal circuit consists of interconnected tubes which has: a) an input section (inflow); b) an output section (outflow); c) an axis of rapid filling up; d) a central body connected with a filter; e) a deflow section; f) a series of multiperforated catheters in their extremities.
6.2.2.2 The priming
The priming, defined as the liquid filling the circuit could be of various type: a) peritoneal dialysis solution [97], physiologic solution [100] or a composition of Normosol solution R pH 7.4 associated with Haemagel (in the proportion 2:1) [117]. The priming volume ought to be abundant enough to achieve homogeneity and constancy of heating, but not excessive, in order to avoid abdominal distension and bodily thermo-dilution. For an optimal working of circuit 3-4 l of perfusate for opened technique and 6 l for closed technique is usually sufficient.
6.2.2.3 The drugs schedule
The drug schedule elected in the current study is Cisplatin (CDDP) (43.0 mg/l of perfusate) and Adriamycin (Dx) (15.25 mg/l of perfusate) [174].

1- 2 - 3 - 4 - 5 - 6 - 7