Neoadjuvant chemotherapy in advanced ovarian cancer: latest results and place in therapy

Abstract

Approximately 70% of women with epithelial ovarian cancer (EOC) are diagnosed with advanced stage disease, which is associated with high morbidity and mortality. The standard approach to treating patients with advanced EOC remains primary debulking surgery (PDS) followed by chemotherapy. EOC is one of the most sensitive of all solid tumors to cytotoxic drugs, with over 80% of women showing a response to standard chemotherapy combined with taxane and platinum. Furthermore, residual disease is a major prognostic factor for survival. On the basis of the clinical features, neoadjuvant chemotherapy (NACT) followed by interval debulking surgery (IDS) is considered to be an alternative treatment option to standard treatment in patients unable to undergo complete resection during PDS. Noninferiority of NACT-IDS to PDS has been demonstrated in some randomized controlled trials and meta-analyses. NACT would also lead to improved quality of life (QOL) of patients, however there are still problems to be solved in the treatment strategy. The uncertainty of perioperative visual assessment of tumor dissemination after NACT has been reported. In addition, several papers have shown the possibility that NACT induces platinum resistance. Furthermore, a notable risk associated with NACT is that patients with significant side effects and refractory disease will lose the opportunity for debulking surgery. Appropriate selection of the patient cohort for NACT is an important issue. Bevacizumab (Bev) is active in patients with advanced EOC. However, the use of Bev is not recommended in the neoadjuvant setting. Bev has a specific adverse event profile that needs to be considered, especially for surgical management, such as gastrointestinal perforation, hemorrhage, and thromboembolic events. NACT could be an alternative treatment option in patients with stage III or IV EOC. However, further studies are needed to clarify the precise role of NACT in the management of advanced EOC.

Introduction

Epithelial ovarian cancer (EOC) is the leading cause of death in women with gynecological malignancy [Ferlay et al. 2010]. The estimated annual incidence of EOC is 225,500 with an estimated 140,200 deaths worldwide in 2008, consisting of 3.7% of all female cancers and 4.2% of cancer deaths [World Health Organization, 2011]. Due to inadequate screening tools and a lack of early clinical symptoms, approximately 70% of women with EOC are diagnosed with advanced stage of disease, which is associated with high morbidity and mortality [Cannistra, 2004; Heintz et al. 2006; Jemal et al. 2010]. Currently, standard primary therapy for patients with advanced EOC is primary debulking surgery (PDS) aiming to remove all visible tumor tissue, followed by adjuvant chemotherapy (ACT) with paclitaxel and carboplatin [Du Bois et al. 2005; Du Bois and Pfisterer, 2005; Pignata et al. 2011]. Despite treatment with this strategy, the majority of these patients develop a relapse within the first 5 years after initial diagnosis and only 20–25% of cases are cured. Furthermore, 5-year survival rate of patients with advanced EOC has not seen a clear improvement in the last decade.

Recently, interval debulking surgery (IDS) after a short course of neoadjuvant chemotherapy (NACT), usually three cycles of chemotherapy, has become a possible alternative treatment option to standard treatment in patients unable to undergo complete resection during PDS. Several randomized trials have shown that, although progression-free survival (PFS) and overall survival (OS) rates in patients given NACT-IDS were not different from those of patients undergoing PDS, patients who received NACT had significantly lower adverse effect and mortality rates after IDS than patients undergoing PDS [Van Der Burg et al. 1995; Rose et al. 2004; Vergote et al. 2010]. Therefore, the significance of NACT-IDS has been further appreciated. In this review, we describe the latest knowledge relating to the use of NACT-IDS to treat advanced EOC.

The place of NACT in the treatment of advanced EOC

Residual disease at the end of surgery is a major prognostic factor for survival [Winter et al. 2007], justifying extensive cytoreductive surgery. Gynecologic oncologists perform a resection of disseminated disease by resecting the peritoneum and other organs such as the intestinal tract, liver and spleen, in addition to staging laparotomy. Over the last decade, the goal of advanced EOC debulking surgery has changed from residual tumor less than 1 cm to no macroscopic residual tumor both in PDS and IDS [Hoskins et al. 1992, 1994; Eisenkop et al. 1998; Vergote et al. 1998, 2010, 2011a; Bristow et al. 2002; Aletti et al. 2006; Chi et al. 2006, 2009; Winter et al. 2007; Du Bois et al. 2009]. However, complete resection of the tumor is often difficult for patients with massively disseminated tumors.

EOC is one of the most sensitive of all solid tumors to cytotoxic drugs, with over 80% of women showing a response to standard chemotherapy combining taxane and platinum. Even if preoperative diagnostic imaging shows massive ascites and diffuse dissemination, these show a dramatic disappearance at IDS after NACT. Based on these clinical characteristics, NACT has been proposed to reduce the burden of disease in patients with bulky disease [Jacob et al. 1991; Schwartz et al. 1994; Vergote et al. 1998, 2000]. While the standard approach to treating patients with advanced EOC remains PDS followed by platinum-based chemotherapy, NACT-IDS is a treatment approach gaining increasing popularity [Schwartz, 2008, 2009; Fago-Olsen et al. 2014].

NACT is defined as the chemotherapy performed prior to cytoreductive surgery. In recent years, NACT-IDS has gained credibility as a valid therapeutic strategy especially for patients with stage IV unresectable bulky tumors or poor general condition [Cannistra, 2004; Rauh-Hain et al. 2012]. NACT setting treatment is now expected to become a standard treatment or one of the effective treatment options for advanced EOC.

Clinical evidence of NACT

To date, there have been some prospective, randomized studies examining the utility of NACT-IDS in patients with advanced EOC (Table 1).

Table 1.

Characteristics of randomized control trials.

Author, year	n (eligible)	FIGO stage	Residual disease status	Interventions	Median OS, months (significance)
Van Der Burg, 1995 (EORTC trial)	278	IIB–IV	> 1 cm	Arm 1: IDS after three cycles of triweekly CPA and CDDP followed by three more cycles of the same regimen	26.0
				Arm 2: PDS followed by six cycles of the same regimen of chemotherapy	20.0 [p = 0.01]
Rose, 2004[GOG trial]	424	III–IV	> 1 cm	Arm 1: IDS after three cycles of triweekly TP, followed by three more cycles of the same regimen	33.9
				Arm 2: PDS followed by six cycles of the same regimen of chemotherapy	33.7 [NS]
Vergote, 2010 (EORTC/NCIC trial)	670	IIIC–IV	> 1 cm	Arm 1: IDS after three cycles of platinum-based NACT, followed by three more cycles of ACT	29.0
				Arm 2: PDS followed by at least six courses of platinum-based chemotherapy	30.0 [NS]

ACT, adjuvant chemotherapy; CDDP, cisplatin; CPA, cyclofosfamide; EORTC, European Organization for the Research and Treatment of Cancer; GOG, Gynecologic Oncology Group; IDS, interval debulking surgery; NACT, neoadjuvant chemotherapy; NCIC, National Cancer Institute of Canada; NS, not significant; OS, overall survival; PDS, primary debulking surgery; TP, paclitaxel/cisplatin.

A randomized phase III trial conducted by the European Organization for the Research and Treatment of Cancer (EORTC) evaluated the benefit of IDS after suboptimal primary debulking by comparing 140 patients who received three cycles of cisplatin and cyclophosphamide chemotherapy followed by IDS and three additional cycles of ACT with 138 similar patients receiving the same chemotherapy regimen without IDS. The IDS group had a statistically significant advantage in median survival time (26 months) compared with patients not undergoing IDS (20 months) [Van Der Burg et al. 1995].

Similarly, in a randomized phase III trial conducted by the Gynecologic Oncology Group (GOG), 550 patients with stage III and IV EOC left with residual disease greater than 1 cm following an initial attempt at PDS [Rose et al. 2004]. All patients received three cycles of initial chemotherapy with cisplatin and paclitaxel followed by response evaluation. Patients whose disease had not progressed during the treatment interval were randomly assigned to IDS plus three additional cycles of ACT or additional chemotherapy alone. In contrast to the EORTC trial, the likelihood of PFS in the group assigned to IDS plus chemotherapy was not significantly different compared with the chemotherapy alone group [hazard ratio (HR) 1.07, 95% confidence interval (CI) 0.87–1.31, p = 0.54]; there was also no significant difference in relative risk of death for patients undergoing interval surgery (relative risk 0.99, 95% CI 0.79–1.24, p = 0.92).

A randomized trial was recently performed by EORTC and the National Cancer Institute of Canada (NCIC) comparing PDS with NACT-IDS [Vergote et al. 2010]. In this trial, 718 patients with epithelial ovarian, fallopian tube or primary peritoneal carcinoma were enrolled. All patients had International Federation of Gynecology and Obstetrics (FIGO) stage IIIC and IV disease and were randomly assigned to PDS followed by platinum-based chemotherapy (PDS group) or NACT followed by IDS (NACT group). The largest residual tumor was 1 cm or smaller (optimal surgery) after PDS in 41.6% of patients and after IDS in 80.6% of patients. Although PFS and OS were similar in both groups, postoperative infections, venous complications, fistula, hemorrhage and postoperative mortality tended to be higher after PDS. A noteworthy drop in OS was noted during the first 3 months after randomization as a result of postoperative mortality and delay of postoperative chemotherapy in patients undergoing PDS. Complete resection of all macroscopic disease (at PDS or IDS) was the strongest independent variable in predicting OS. The authors concluded that NACT-IDS was not inferior to PDS followed by chemotherapy as a treatment option for patients with bulky stage IIIC or IV EOC and complete resection of all macroscopic disease, whether performed as primary treatment or after NACT, remains the objective whenever cytoreductive surgery is performed.

Several other phase III trials are ongoing aiming to address the question of whether NACT before surgery could replace the primary surgery before ACT in terms of survival comparison (Table 2). The latest results of the trial run by the Royal College of Obstetricians and Gynecologists in the UK (CHORUS study) were presented at the 2013 American Society of Clinical Oncology annual meeting [Kehoe et al. 2013]. This trial was designed to demonstrate the noninferiority of NACT to PDS. A total of 552 patients with stage III/IV EOC were enrolled between 2004 and 2010. Of these, 276 had been randomly assigned to PDS followed by six cycles of platinum-based ACT and 274 had been randomly assigned to three cycles of platinum-based NACT followed by surgery and then three cycles of ACT. Baseline characteristics were well balanced. Median age was 65.5 years, median tumor size was 8 cm and 25% had stage IV disease. Median duration of follow up was 3 years. About 20% of patients in both arms had poor performance status (PS). Optimal debulking was possible in 16% of the PDS arm versus 40% of the NACT arm. Grade 3 or higher toxicity occurred in 48% and 40% in the NACT and PDS groups respectively, while postoperative complications of grade 3 or 4 occurred in 24% and 14% respectively. Fewer deaths within 28 days were reported with NACT: 14 (5.6%) deaths were noted in the PDS arm while 1 (0.5%) occurred in the NACT arm. Intention to treat analysis showed a median OS of 22.8 months and 24.5 months for PDS and NACT respectively (HR 0.87 in favor of NACT, 80% CI 0.76–0.98) and median PFS of 10.2 and 11.7 months respectively (HR 0.91, 80% CI 0.81–1.02) [Kehoe et al. 2013]. These results strengthened the evidence that NACT-IDS is not inferior to PDS.

Table 2.

Characteristics of ongoing randomized control trials.

Study	CHORUS	Kumar	JCOG0602
Identifier	NCT00075712	NCT00715286	UMIN000000523
Timing of randomization	Randomized at entry	Randomized at surgery	Randomized at entry
Number randomized (estimated)	150	180	300
Participants	Stage: III–IV	Stage: IIIc–IV (pleural effusion only)	Stage: III–IV
	Age < 18	Age: 25–65	Age: 20–75
	Pelvic mass with extrapelvic metastases	PS ECOG: 0–2	PS ECOG: 0–3
	Serum CA125/CEA > 25	Cytology/biopsy positive	Cytology positive
			Serum CA125 > 200 U/ml and CEA <20 ng/ml
Interventions	Arm 1: PDS followed by six cycles of triweekly TC or CBDCA alone. Patients may undergo IDS after three courses of chemotherapy	Arm 1: PDS followed by ACT	Arm 1: PDS followed by eight courses of triweekly TC. Patients may undergo IDS after four courses of chemotherapy
	Arm 2: three courses of chemotherapy as in arm 1 followed by IDS, and three courses of ACT	Arm 2: NACT followed by IDS	Arm 2: four courses of chemotherapy as in arm 1 followed by IDS, and four courses of ACT
Outcomes of interest	OS, PFS	Optimal debulking rate	OS, PFS
	QOL	OS, DFS, QOL	Response rate, adverse effects, perioperative complications

ACT, adjuvant chemotherapy; CBDCA, carboplatin; DFS, disease-free survival; ECOG, Eastern Cooperative Oncology Group;; IDS, interval debulking surgery; JCOG, Japan Clinical Oncology Group; NACT, neoadjuvant chemotherapy; OS, overall survival; PDS, primary debulking surgery; PFS, progression-free survival; PS, performance status; QOL, quality of life; TC, paclitaxel/carboplatin.

The Japan Clinical Oncology Group in Japan (JCOG 0602) designed a similar trial (Table 2), with patients in the NAC arm having IDS (which is actually the first or primary surgery) before further chemotherapy, while patients in the other arm had conventional treatment (patients in the conventional arm may also have had IDS upon direction of the physician). To directly address the definite role of IDS, future studies should focus on a comparison of current conventional primary surgery with ACT versus NACT without any attempt to remove the bulk of tumors followed by IDS (which should be allowed only in this arm) and then by further chemotherapy. The appropriate number of cycles of NACT remains uncertain. The safety and efficacy of six cycles of paclitaxel/carboplatin (TC) therapy as NACT has been confirmed [Da Costa Miranda et al. 2014; Stoeckle et al. 2014]. However, the duration of the initial treatment will be longer when ACT is added after NACT-IDS [Milam et al. 2011]. Therefore, a study of the optimal number of cycles of NACT is necessary.

In a meta-analysis including articles between 1989 and 2005, a total of 835 patients with stage III and IV EOC treated with platinum-based NACT in lieu of PDS were evaluated with regard to OS and relative effect of multiple prognostic variables [Bristow and Chi, 2006]. In this study, median survival time was 24.5 months and optimal surgery occurred in 65% of patients. Median OS was positively correlated with platinum plus taxane chemotherapy and more recent year of publication, and negatively correlated with the proportion of stage IV disease. A prognosis improvement of 1.9 months was observed for every percentage of optimal cases increase 10%. However, shortened survival of 4.1 months was observed in each increased by one course of NACT. The median survival time of the NACT group was equivalent to that of patients with PDS ended in suboptimal in GOG study (24 months versus 24.5 months).

Meanwhile, a similar meta-analysis of 21 studies between 1989 and 2008 was performed [Kang and Nam, 2009]. In this study, medium survival time was 27.5 months and optimal cytoreduction rate was 70%. Increasing median OS time (MST) was observed: with more recent year of publication, with increased percentage of taxane use, and increased rate of optimal cytoreduction. However, the number of NACT cycles before IDS and the proportion of patients with stage IV disease did not affect MST.

The latest retrospective studies verified the clinical significance of NACT-IDS. A Danish group compared the outcomes of NACT-IDS (n = 515) with PDS (n = 990). No difference in median OS was observed between PDS and NACT-IDS. However, patients without residual tumor had a better median OS when treated with PDS. In a multivariate analysis, NACT-IDS was associated with an increased risk of death after 2 years of follow up (HR 1.81, CI 1.39–2.35) [Fago-Olsen et al. 2014].

Based on these findings, NACT-IDS has become a primary treatment for patients with advanced EOC [Vergote et al. 2011b; Cornelis et al. 2012]. However, despite NACT being useful for patients in whom optimal debulking appears impossible, primary surgical cytoreduction should not be precluded by a lack of surgical skills and experience [Chi et al. 2012; Vergote et al. 2013].

Evaluation of effect of NACT

EOC staging is surgical and based on laparotomy findings with histological confirmation [Benedet et al. 2000]. Visual estimation by the surgeon is critical for the evaluation of intra-abdominal tumor spread. Whether the surgeons’ statement of complete tumor resection is equal in primary surgery and in IDS remains unclear. NACT before surgery can cause fibrosis and adhesions in the peritoneal cavity and may interfere with the perioperative evaluation of tumor spread. Recently, a paper on the uncertainty of perioperative visual assessment of tumor dissemination after NACT was reported [Hynninen et al. 2013]. In this study, systematic visual evaluation of tumor spread was performed at the start of primary surgery/diagnostic laparotomy (n = 39) or interval surgery (n = 16). The peritoneal cavity was divided into 22 anatomical regions. The carefully documented results of the visual assessment were compared with the histopathological analysis of 220 biopsies from primary and 92 biopsies from interval surgery. In primary surgery, perioperative visual estimation of tumor spread showed 98% sensitivity, 76% specificity and 95% accuracy compared with histopathology. The difference in sensitivity and accuracy in primary and interval operations was statistically significant (p < 0.001). The authors concluded that in advanced EOC, microscopically carcinomatous areas have a benign visual appearance more often after NACT than at primary surgery. NACT may interfere with the perioperative visual evaluation of tumor spread and thus lead to incomplete resection of tumor in potentially resectable areas.

In the histopathological assessment, tumor response to NACT has been reported to cause observable microscopic changes such as tumor necrosis, fibrosis, macrophage infiltration and tumor-induced inflammation [Le et al. 2007; Wang and Zheng, 2013]. These variables have been shown to be significant prognostic factors in other solid tumors and may also be helpful in EOC treatment planning. Pathological assessment of 101 patients with EOC after NACT found that a high pathological tumor response score was the only significant predictor of time to disease-related death [Le et al. 2007]. Moreover, pathological features after NACT, such as fibrosis and necrosis, have been shown to affect outcomes in patients with EOC [Samrao et al. 2012]. Other researchers have shown the utility of histological assessment of surgical specimens after NACT [Muraji et al. 2013]. Outcomes were evaluated retrospectively in patients with advanced EOC or peritoneal cancer who received NACT consisting of paclitaxel and carboplatin followed by IDS. Therapeutic response was assessed histopathologically as grade 0–3, based on the degree of disappearance of cancer cells, displacement by necrotic and fibrotic tissue, and tumor-induced inflammation. Multivariate analysis showed that stage IV disease, residual cancer at the end of surgery of at least 1 cm, and histological grade 0–1 were independent predictors of decreased OS. Grade 0–1 was also an independent predictor of increased risk of relapse within 6 months.

As noted above, visual evaluation of NACT is difficult. Assessing the therapeutic effects of NACT by histological specimens may be important in the choice of drug to be used in ACT.

Platinum resistance

Drug resistance after NACT and ACT has been found to correlate with in vitro drug resistance [Lim et al. 2010]. Several papers have shown the possibility of NACT inducing platinum resistance [Matsuo et al. 2010b; Chi et al. 2012]. The EORTC-NCIC randomized trial showed that surgery was less extensive in patients who received NACT, with associated reductions in mortality and postoperative complication rates. Nevertheless, NACT did not extend OS, perhaps due to tumor development of drug resistance.

The latest study comparing the response of chemotherapy in the PDS group and the NACT-IDS group was reported by Rauh-Hain and colleagues [Rauh-Hain et al. 2013]. The study population consisted of 425 patients, 95 (22.3%) underwent NACT-IDS and 330 (77.6%) underwent PDS. After the initial platinum-based chemotherapy, 42 (44.2%) women in the NACT-IDS group were considered to have platinum-resistant disease compared with 103 (31.2%) in the PDS group (p = 0.01). When multivariate logistic regression was used to control for factors independently associated with platinum resistance, NACT-IDS was no longer associated with an initial increased risk. However, in women who had a recurrence and were retreated with platinum-based chemotherapy, 32 (88.8%) in the NACT-IDS group had a recurrence within 6 months and were considered platinum resistant compared with 62 (55.3%) in the PDS (p < 0.001). The authors concluded that in women with EOC who have a recurrence and are treated again with platinum-based chemotherapy, NACT-IDS appears to increase the risk of platinum resistance. The data to support the clinical facts as described above are not clear. However, the possibility that ovarian cancer stem cells have remained in the abdominal cavity after NACT has been reported. After IDS, in which complete removal of macroscopic tumor tissue is achieved, residual cancer stem cells remain the in scar tissue. A population of chemotherapy-resistant stem cells selected during NACT and not debulked in IDS may play a role in EOC recurrence. It was suggested that all tissue showing traces of tumor, that is, scar tissue, should be removed during IDS [Lim et al. 2010]. More recently, another author reported that TP53-K351N mutation was involved in platinum resistance after NACT [Zhang et al. 2014]. However, at present, the possibility of NACT inducing chemotherapy resistance is unclear and further study is needed.

Recently, selection of chemotherapy regimens based on histology have attracted attention. The response rate of TC therapy for clear cell adenocarcinoma (CCC) and mucinous adenocarcinoma (MAC) is low compared with endometrioid adenocarcinoma and serous adenocarcinoma [Goff et al. 1996; Sugiyama et al. 2002; Shimada et al. 2009]. Efficacy of CPT-11 (CPT) against CCC has been confirmed in vitro and in vivo [Itamochi et al. 2002]. The Japanese Gynecologic Oncology Group (JGOG) carried out a prospective randomized phase II trial comparing CPT-cisplatin (P) therapy and TC therapy as initial chemotherapy for stage IC–VI CCC. Because, a favorable trend was observed for PFS in the CPT-P therapy group, an international randomized controlled phase III trial comparing CPT-P with TC is underway (GCIG/JGOG3017). It may be that CPT-P becomes an effective regimen of NACT for CCC. However, for MAC, the effectiveness of agents such as CPT, 5 fluorouracil and oxaliplatin used in gastrointestinal cancer has been studied [Sato et al. 2009]. Because an effective regimen has not been determined, NACT is not recommended for MAC. In addition, the effectiveness of NACT-IDS for nonepithelial ovarian tumors such as malignant germ cell tumor and yolk sac tumor has been reported [Lu et al. 2014; Talukdar et al. 2014].

The benefit of NACT relies on the correct selection of effective chemotherapy regimens. An assessment of the individual patient’s chemosensitivity is essential for providing effective chemotherapy. In recent years, several biomarkers and methods for predicting the response to chemotherapy have been investigated [Kawaguchi et al. 2005; Naniwa et al. 2007; Matsuo et al. 2010a] but never used widely. Specific biomarkers need to be determined to identify patients most likely to benefit from NACT. Recently, new findings on clinical biomarkers useful in treatment selection (PDS or NACT-IDS) for advanced ovarian cancer have been reported. Exploratory post hoc analyses of registered cases in the EORTC55971 trial were performed. They found that the size of the largest metastatic tumor and clinical stage were significantly associated with the patient’s prognosis. More specifically, patients with stage IIIC disease and metastatic tumors up to 45 mm benefited more from primary surgery while those with stage IV disease and metastatic tumors larger than 45 mm benefited more from NACT [Van Meurs et al. 2013]. The authors concluded that both treatment options led to comparable survival rates for patients who did not meet these criteria. Further clarification of the treatment selection rule is necessary for progress towards individualized medicine.

Older patients

Older women with EOC are less likely to receive care from a gynecologic oncologist, undergo aggressive cytoreductive surgery and less likely to receive platinum-based chemotherapy or clinical trial participation [Hershman et al. 2004; Wright et al. 2008]. Oncologic outcomes among older women undergoing NACT-IDS appear similar to those undergoing PDS. The risk of readmission within 30 days of surgery was significantly greater among patients undergoing PDS compared with IDS. The increased risk of hospital readmission after PDS should be considered when contemplating NACT-IDS versus PDS as primary treatment [Worley et al. 2013]. Another study revealed that patients aged 70 years and over who underwent NACT had less perioperative morbidity after IDS, had improved complete cytoreduction to no residual disease (71.4% versus 28.1%), and there was no difference in OS or PFS [Glasgow et al. 2013]. A retrospective study evaluated use of NACT in patients aged 65 years and over. In their retrospective cohort, 20% of patients received NACT and there was no difference in rate of complications, rate of complete surgical resection or difference in OS and PFS [McLean et al. 2010].

A retrospective study from Memorial Sloan-Kettering found that 10% of their patient population received NACT because of advanced stage, medical comorbidities or advanced age (> 85 years). They found a statistically significant difference in OS (37 versus 50 months) and in PFS (13 versus 17 months) in patients treated with NACT versus PDS. They observed that in women who had complete surgical cytoreduction to no residual disease, the PFS was 24 months, which far exceeds the PFS of patients in the EORTC study (11 months). The authors concluded that primary debulking should continue to be the preferred management of patients with stage IIIC-IV EOC [Chi et al. 2012].

Wright and colleagues performed a population-based analysis to examine the effectiveness of upfront treatment strategies in 9587 older women (> 65 years old) with stage II–IV EOC. They found that use of PDS decreased from 63.2% in 1991 to 49.5% by 2007, whereas NACT increased from 19.7% in 1991 to 31.8% in 2007. Furthermore, in the observational cohort, survival with NACT did not differ significantly from that of PDS [Wright et al. 2014]. The importance of NACT-IDS in the initial treatment for older women with EOC will increase in the future.

Quality of life

NACT is thought to lead to improved QOL of patients. Patients with advanced disease frequently experience a variety of treatment- and disease-related side effects which may diminish their QOL. Patient-reported QOL has been recommended as an endpoint in clinical trials. The National Cancer Institute and the Food and Drug Administration mandated that the treatment goals should not only focus on survival but also on QOL. Numerous clinical trial protocols have included QOL as a secondary endpoint but until now only a few publications reported QOL outcomes in phase III ovarian cancer trials [Bezjak et al. 2004; Wenzel et al. 2005; Greimel et al. 2006; Rustin et al. 2011].

The EORTC trial included QOL as a secondary endpoint [Vergote et al. 2010]. Survival and QOL after NACT followed by surgery was similar to that after PDS followed by chemotherapy. However, institutions with good QOL compliance had a higher optimal debulking rate and better survival outcomes [Greimel et al. 2013]. Schwartz and colleagues reported that the NACT group had a poor PS and were significantly older compared with the PDS group in stage IIIC and IV cases; however, the length of hospital stay was significantly shorter in the IDS group [Schwartz et al. 1999].

These observations suggest that NACT plays an important role in maintaining QOL of patients with advanced EOC. Because advanced EOC is a disease that can rarely be cured, QOL should be evaluated as an endpoint in clinical trials of NACT.

Bevacizumab

Bevacizumab (Bev) is a humanized monoclonal antibody that recognizes circulating vascular endothelial growth factor. Several studies have reported that Bev as first- or second-line treatment is active in patients with advanced EOC [Cannistra et al. 2007; Burger et al. 2011; Perren et al. 2011; Sato and Itamochi, 2012]. However, the use of Bev is not yet recommended in the neoadjuvant setting for the management of advanced EOC. This is because potentially problematic adverse effects unique to its mechanism of action could occur, which is especially important for surgical management. The most serious adverse events are gastrointestinal perforations, hemorrhages and arterial thrombolic events such as stroke and myocardial infarction. Gastrointestinal perforations associated with Bev used in the perioperative period have attracted attention because they seem to be more common in EOC than in other solid tumors [Han and Monk, 2007].

In the study by Cannistra and colleagues [Cannistra et al. 2007], patients were excluded if they had undergone major surgical procedure within 28 days. Despite the relatively good condition of the patients regarding adverse effects of Bev, the sponsor closed the study early because of a higher than expected incidence of gastrointestinal perforation (11.4%). In another study [Garcia et al. 2008], the exclusion criteria included serious, nonhealing wound ulcers and major surgical procedure. Among the 70 patients entered, three patients developed a gastrointestinal perforation, one a gastrointestinal fistula and another patient had a wound-healing complication. However, no cases of gastrointestinal perforations have been reported in other phase II trials evaluating Bev in the frontline setting [Burger et al. 2011]. Recently, Chereau and colleagues evaluated the safety and postoperative course of patients who had received Bev before debulking surgery for advanced EOC. They described the case of five patients who were initially judged to have inoperable disease and were revaluated after six courses of chemotherapy. The rate of postoperative complications was high (four among the five patients), but only one patient had a grade 3 or higher complication. None of the patients died in the postoperative course [Chereau et al. 2013]. Further investigation is necessary in order to confirm the usefulness of using Bev for NACT.

Conclusion

Noninferiority of NACT-IDS to PDS has been demonstrated in randomized controlled trials. However, there is still no evidence that NACT is superior to standard treatment. The biggest risk associated with use of NACT is that patients with significant side effects and refractory disease will lose the opportunity for initial surgery. Establishment of an optimal regimen is necessary in order to improve the outcome of NACT. Furthermore, the precise role of NACT in the management of advanced EOC has not yet been established. Importantly, well designed clinical trials of NACT are essential. Further studies to clarify the role of molecular targeted therapies in NACT are required.