Indications for Laparoscopic Assessment of Cytoreduction

Background

The current standard of care for advanced epithelial ovarian cancer (AEOC) is a combination of cytoreductive surgery and paclitaxel and platinum-based chemotherapy. The size of residual disease has been repeatedly proven to be a crucial prognostic factor for affected patients, and the achievement of complete gross tumor resection is now recognized as the main goal of cytoreductive surgery. In a meta-analysis by Chang and colleagues, the authors sought to quantify the impact of complete cytoreduction to no gross residual disease on overall survival among patients with advanced-stage ovarian cancer. A total of 13,257 patients were included in the analysis. The authors found that after controlling for other factors with multiple linear regression analysis, each 10% increase in the proportion of patients undergoing complete cytoreduction to no gross residual disease was associated with a significant and independent 2.3-month increase in median survival compared with a 1.8-month increase in cohort median survival for optimal cytoreduction (residual disease <1 cm). It has also been previously shown that optimal cytoreduction (≤1 cm residual disease) may be achieved in up to 71% of patients.

Several factors may lead to a suboptimal primary cytoreduction. These include poor patient selection, lack of surgeon expertise in radical abdominal and pelvic surgery, lack of availability of consultants to achieve complete tumor resection, and failure to accept the principle that optimal cytoreduction may offer an advantage to patients in terms of oncologic outcomes. Another important factor when considering the impact of the initial surgical approach is the fact that the rate of complications after up-front surgery is not insignificant. This is of relevance, given that one should aim to minimize exposure of the patient to such complications if there is not going to be a benefit derived from the operation. Therefore it is imperative to identify tools that will ultimately allow surgeons to obtain the information required to make an adequate decision regarding which patients are ideal candidates for surgical treatment and which are ideal candidates for neoadjuvant chemotherapy.

Imaging Tools for Preoperative Evaluation

A number of variables have been previously used to determine when to proceed with primary cytoreduction. These include computed tomography (CT) scans, serum CA-125 levels, and overall patient performance status.

Serum Biomarkers

The most frequently used serum biomarker for ovarian cancer is CA-125. In prior studies it was proposed as a predictor of suboptimal cytoreduction. In a publication by Chi and colleagues, the investigators evaluated 100 consecutive patients with stage III ovarian cancer. Optimal cytoreduction (≤1 cm) was achieved in 73% of patients with a CA-125 level below 500 U/mL, compared with only 22% of those with a CA-125 level above 500 U/mL ( P < .001). In a subsequent study, the same investigators reported evaluating preoperative CA-125 levels as a predictor of outcome of primary debulking surgery incorporating the element of upper abdominal surgery. In that study, the authors found that 25% of patients had R0 resection, 55% had residual cancer less than 1 cm, and 20% had residual cancer larger than 1 cm. They concluded that there was no threshold CA-125 level that accurately predicted cytoreductive outcomes. A subsequent meta-analysis by Kang and colleagues evaluated 14 studies with 2192 patients to assess the performance of CA-125 at various cutoff levels as a predictor of the outcome of a cytoreductive surgical procedure. The authors found that preoperative serum CA-125 level higher than 500 U/mL was strongly associated with suboptimal cytoreduction.

Preoperative Imaging Modalities

The most commonly used imaging modality when evaluating patients with advanced-stage ovarian cancer is CT. Several models have been proposed, and these are reviewed in Table 11.1 . Nelson and colleagues scored CT scans, on the basis of criteria presented in Table 11.1 , as cytoreducible (no disease remaining in criteria site) or not cytoreducible (at least one site of disease remaining) by standard surgical techniques. Optimal cytoreduction was defined as less than 2 cm, and this was achieved in 23 of 24 patients with disease scored as cytoreducible and in 6 of 18 patients with disease scored as not cytoreducible. The authors noted that the CT scans predicted surgical outcome with a sensitivity of 92.3% and a specificity of 79.3%. A subsequent study by Bristow and colleagues proposed another CT-based predictive model. Thirteen radiographic features met the inclusion criteria and were assigned 1 or 2 points, and a Gynecologic Oncology Group performance status score of 2 or higher (assigned 2 points) was used to calculate a predictive index score. The authors reported that a predictive index score of 4 or higher had the highest overall accuracy, at 92.7%, and identified patients undergoing suboptimal cytoreduction with a sensitivity of 100%. Later, Dowdy and colleagues published a retrospective analysis reviewing 87 preoperative CT scans for 17 criteria that outlined extent of disease. In that study, the authors found that a model based on diffuse peritoneal thickening and ascites had a 68% positive predictive value and 52% sensitivity and was associated with a low rate of optimal cytoreduction (32%).

Table 11.1

Computed Tomography (CT)-Based Models to Predict Likelihood of Optimal Cytoreduction in Advanced Ovarian Cancer

Model	CT Findings
Nelson	Attachment of the omentum to the spleen Disease >2 cm in: Mesentery Liver surface or parenchyma Diaphragm Gallbladder fossa Suprarenal paraaortic nodes Pericardial nodes Pulmonary or pleural nodules
Bristow	2 points for each of the following: Peritoneal thickening Peritoneal implants ≥2 cm Small bowel mesentery disease ≥2 cm Large bowel mesentery disease ≥2 cm Omental disease extension to the stomach, spleen, or lesser sac Extension to pelvic sidewall, parametria, or hydroureter Large-volume ascites (seen on all cuts) Suprarenal paraaortic lymph nodes ≥1 cm 1 point for each of the following: Diaphragm or lung disease ≥2 cm or confluent plaque Inguinal canal disease or lymph nodes ≥2 cm Liver lesion ≥2 cm on surface or parenchymal lesion of any size Portahepatic or gallbladder fossa disease ≥1 cm Infrarenal paraaortic lymph nodes ≥2 cm
Dowdy	Diffuse peritoneal thickening, defined as peritoneum >4 mm thick in at least 2 of the following 5 areas: Lateral colic gutters Lateral conal fascia Anterior abdominal wall Diaphragm Pelvic peritoneal reflections Ascites present on at least two-thirds of the CT cuts

Combination Predictive Models

More recently, a number of investigators have explored whether a multimodality evaluation of patients with advanced ovarian cancer can predict no gross residual disease with a high level of accuracy. In a recent study by Suidan and colleagues, the authors sought to assess the ability of preoperative CT scans and CA-125 to predict no gross residual disease at primary cytoreduction in advanced ovarian cancer. That study was a secondary post hoc analysis of a previously published prospective, nonrandomized, multicenter trial that had identified nine criteria for suboptimal (>1 cm residual) disease. Four clinical and 18 radiologic criteria were assessed, and a multivariate model predictive of residual disease was developed. The investigators found that on multivariate analysis, three clinical and eight radiologic criteria were significantly associated with the presence of any residual disease: age 60 years or older (odds ratio [OR], 1.5); CA-125 level 600 U/mL or higher (OR, 1.3); American Society of Anesthesiologists (ASA) classification category 3 or 4 (OR, 1.6); lesions in the root of the superior mesenteric artery (OR, 4.1); splenic hilum or ligaments (OR, 1.4), lesser sac greater than 1 cm (OR, 2.2), gastrohepatic ligament or porta hepatis (OR, 1.4), gallbladder fossa or intersegmental fissure (OR, 2); suprarenal retroperitoneal lymph nodes (OR, 1.3); small bowel adhesions or thickening (OR, 1.1); and moderate to severe ascites (OR, 2.2). All ORs were significant ( P < .01). A predictive score was assigned to each criterion based on its multivariate OR, and the rate of having any residual disease for patients who had a score of 0 to 2, 3 to 5, 6 to 8, or 9 or higher was 45%, 68%, 87%, 96%, respectively. This study demonstrated a predictive model of 11 criteria that were associated with residual disease. In other words, this was a predictive model in which the rate of having any residual disease was directly proportional to a predictive score.

Another potential imaging modality of patients with advanced ovarian cancer is the combination of contrast-enhanced diffusion-weighted magnetic resonance imaging (DW-MRI). Recently, a group of investigators evaluated DW-MRI for staging and assessing operability compared with CT and positron emission tomography–computed tomography (PET-CT) in patients with suspected ovarian cancer. The authors found that DW-MRI showed 94% accuracy for primary tumor characteristic compared with 88% for CT and 94% for PET-CT. DW-MRI showed a higher accuracy of 91% for peritoneal staging compared with CT (75%) and PET-CT (71%). The authors concluded that DW-MRI showed high accuracy for characterizing primary tumors and peritoneal and distant staging compared with CT and PET-CT.

Despite numerous predictive clinical, radiologic, or serologic variables, there are no conclusive models that provide objective and accurate measures that serve as predictors of microscopic residual disease. In addition, it has been challenging to demonstrate consistent reproducibility of such models outside of the institutions that have created such models. In addition, there is a significant element of variability regarding the surgical aggressiveness in the respective institutions where such models originate.

Rationale for Laparoscopic Evaluation

A laparoscopic evaluation of the abdomen and pelvis to determine whether the tumor volume is resectable might offer surgeons an additional tool helping them decide who is a good candidate for up-front cytoreduction. In addition, the laparoscopic approach allows for a fast recovery of the patient, so that if it is decided not to proceed with an attempt at initial cytoreduction, the patient will then be able to undergo treatment in the form of systemic chemotherapy much sooner. The laparoscopic approach offers excellent visualization of all quadrants of the abdomen and pelvis. In addition, it allows for tissue biopsy for definitive diagnosis and molecular analysis. This access to tissue analysis may allow patients to be involved in innovative trials known as window-of-opportunity trials (WOTs). This provides for an evaluation of the clinical and molecular impact of novel therapeutic agents among those triaged to primary cytoreductive surgery. Lastly, the tumor collected at the time of interval cytoreductive surgery can be evaluated for the impact of novel combinations of neoadjuvant therapeutic agents.

History of Laparoscopy to Assess Feasibility of Cytoreduction

Various scores have been proposed, but only one has been uniformly validated and adopted. In 2005, Fagotti and colleagues reported on 65 patients who underwent preoperative clinical-radiologic evaluation followed by laparoscopy and then laparotomy. The investigators evaluated for several elements, including ovarian masses (unilateral vs. bilateral), omental cake or nodules, peritoneal and diaphragmatic carcinomatosis, mesenteric retraction, bowel and stomach infiltration, liver metastases, and bulky lymph nodes. In that study, optimal debulking was achieved in 87% of patients whose disease was deemed resectable on the basis of laparoscopy findings. The overall accuracy rate of laparoscopy in predicting optimal cytoreduction was 90%. In 2006, the same group proposed a laparoscopy-based predictive index value (PIV) based on objective parameters determined at pre-cytoreduction laparoscopy. The findings at laparoscopy were used to estimate the chances of optimal cytoreduction (residual disease ≤1 cm) ( Table 11.2 ). The scores for each item were added to obtain an overall score, the PIV. The authors in that study found that the overall accuracy of the model in predicting surgical outcome was approximately 75% and also that if the PIV was 8 or higher, then the likelihood that a patient would have a suboptimal surgical result was 100%. This was a landmark study because it showed that for the first time laparoscopy rather than imaging studies could predict with excellent accuracy the chance of an optimal cytoreduction.

Table 11.2

Laparoscopic Features and Corresponding Score for Tumor Volume Assessment

Laparoscopic Feature	Score 0	Score 2
Peritoneal carcinomatosis	Carcinomatosis involving a limited area (along the paracolic gutter or the pelvic peritoneum) and surgically removable by peritonectomy	Unresectable massive peritoneal involvement as well as with a miliary pattern of distribution
Diaphragmatic disease	No infiltrating carcinomatosis and no nodules confluent in most of the diaphragmatic surface	Widespread infiltrating carcinomatosis or nodules confluent in most of the diaphragmatic surface
Mesenteric disease	No large infiltrating nodules and no involvement of the root of the mesentery as would be indicated by limited movement of the various intestinal segments	Large infiltrating nodules or involvement of the root of the mesentery indicated by limited movement of the various intestinal segments
Omental disease	No tumor diffusion observed along the omentum up to the large stomach curvature	Tumor diffusion observed along the omentum up to the large stomach curvature
Bowel infiltration	No bowel resection was assumed and no miliary carcinomatosis on the serosa observed	Bowel resection assumed or miliary carcinomatosis on the serosa observed
Stomach infiltration	No obvious neoplastic involvement of the gastric wall	Obvious neoplastic involvement of the gastric wall
Liver metastases	No surface lesion	Any surface lesion

In 2011, Fagotti and her group performed a prospective evaluation of the learning curve for surgeons determining the PIV. The authors compared the scores for each laparoscopic parameter assigned by fellows and senior surgeons. Ninety consecutive women with suspected advanced ovarian or peritoneal cancer underwent laparoscopy performed by a fellow and senior surgeon sequentially with independent assignment of scores for disease distribution. The median PIV was 6 (range, 0–10) for the fellows and 6 (range, 0–14) for the senior surgeons. The authors concluded that fellows in gynecologic oncology with at least 12 months’ experience assigned laparoscopy-based scores similar to those of senior surgeons.

Subsequently, Fagotti and colleagues evaluated the reproducibility of a laparoscopic assessment before cytoreductive surgery. They performed a prospective, multicenter trial (Olympia-MITO 13) in which the laparoscopy-based PIV was evaluated in 10 satellite centers. A total of 120 patients with advanced ovarian, fallopian tube, or primary peritoneal cancer underwent laparoscopy at the satellite centers; the procedures were recorded and blindly reviewed at the coordinator center. The investigators reported an accuracy rate of 80% or greater in 9 of 10 satellite centers.

The original studies on the evaluation of a laparoscopic approach with regard to feasibility of cytoreduction were all based and designed on a definition of optimal residual disease as 1 cm or less. However, in subsequent years the routine use of radical upper abdominal surgical procedures during cytoreductive surgery has significantly increased the chance of achieving complete tumor debulking to microscopic residual disease. To this end, efforts were made to ensure that laparoscopy was still a useful tool in assessing the abdomen and pelvis as a predictor of optimal disease. A subsequent proposal was made wherein mesenteric retraction and miliary carcinomatosis on the serosa of the small bowel were removed from the scoring system. In their study, Petrillo and colleagues sought to update their previous laparoscopy-based model to predict incomplete cytoreduction (residual disease >0) in advanced ovarian cancer after the introduction of upper abdominal surgery. All patients underwent a laparotomy after diagnostic laparoscopy. The presence of omental cake; peritoneal extensive carcinomatosis; diaphragmatic confluent carcinomatosis; bowel infiltration; stomach, spleen, and/or lesser omentum infiltration; and superficial liver metastases was evaluated. No gross residual disease was achieved in 58% of patients. Upper abdominal surgery was required in 53.3% of cases. The authors noted a very high overall agreement between laparoscopy and laparotomic findings, which ranged from 74.7% for omental cake to 94.8% for stomach infiltration. At a laparoscopic PIV of 10 or higher, the chance of achieving complete primary debulking with a surgical procedure was 0, and the risk of unnecessary laparotomy was 33.2%. This new model led to the proposal that a PIV of 10 or higher should be the new determinant to proceed with neoadjuvant chemotherapy, rather than 8 or higher.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here