AGO Study

A large, retrospective review of cytoreductive surgery in 267 women with recurrent ovarian cancer treated in 25 European centers from 2000 to 2003 was recently reported by the Arbeitsgemeinschaft Gynaekologische Onkologie (AGO).[28] This study included patients with stage I (46, 18.0%), stage II (33, 12.9%), stage III (165; 64.7%), and stage IV (11, 4.3%) disease. Patients had had treatment-free intervals of < 6 months (36, 13.5%), 6 to 12 months (63, 23.6%), and > 12 months (168, 62.9%). Ages ranged from 24 to 84 years (median = 60 years), and 91.9% had an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1.

Patients with no macroscopic disease following secondary surgical cytoreduction had a median survival of 45.2 months, while those with macroscopic disease had a median survival of 19.7 months (hazard ratio [HR] = 3.71; 95% confidence interval [CI] = 2.27–6.05; P < .0001; see Figure 3). Interestingly, the size of the macroscopic residual tumor had no impact on survival. The median survival of patients with a residual tumor of 0.1 to 1.0 cm was 19.6 months, and for those with macroscopic residual disease > 1 cm, it was 19.7 months (HR = 0.84, 95% CI = 0.51–1.40, P = .502). In a multivariate analysis, the three factors that affected survival after secondary cytoreduction were complete resection (residual tumor 0 vs > 0 mm: HR = 2.94; 95% CI = 1.68–5.17; P < .001), ascites (< 500 vs ≤ 500 mL: HR = 2.30, 95% CI = 1.31–4.04; P = .004), and platinum-containing chemotherapy (yes vs no: HR = 1.84; 95% CI = 1.13–3.01; P = .015).

Table 4
Univariate Analysis of Significant Factors for Achieving Complete Resection in Ovarian Cancer Patients
	Status	N	P Value	OR	95% CI
ECOG performance status	0 > 0 a	118 149	< .0001	1 2.74	1.66–4.51
FIGO stage	I/II III/IV a	79 188	.01	1	1.18–3.46
Residual disease after primary surgery	0 mm > 0 mm	124 143	.0005	1 2.39	1.46–3.91
Preoperative serum CA-125 b	0-70 U/mL 71–350 U/mL > 350 U/mL	100 102 47	.001	1 1.23 3.76	0.70–2.15 1.77–7.99
Ascites in preoperative diagnostic imaging	< 500 mL ≥ 500 mL	231 36	< .001	1 6.11	2.45–15.23
Localization of recurrence in preoperative diagnostic imaging	Pelvis Others a	71 196	.017	1 1.96	1.12–3.41
Peritoneal carcinomatosis in preoperative diagnostic imaging b	No a Yes	209 58	.0001	1 3.34	1.77–6.31
Intraoperative peritoneal carcinomatosis	No Yes	125 125	< .0001	1 6.87	4.00–11.76
a Missing data were added to this group. b Cancer antigen (CA)-125 and peritoneal carcinomatosis in preoperative diagnostics not calculated in multivariate analysis because of correlation with ascites. CI = confidence interval; ECOG = Eastern Cooperative Oncology Group; FIGO = International Federation of Gynecology and Obstetrics; HR = hazard ratio; OR = odds ratio. Source: Harter P et al.[28]

Thus, as with surgery for previously untreated ovarian cancer, leaving no macroscopic residual disease was the most important factor in prolonging survival for women with recurrent disease. A univariate analysis of significant factors for achieving complete resection are presented in Table 4. A multivariate analysis for achieving complete resection of recurrent disease is presented in Table 5.

Table 5
Multivariate Analysis of Factors for Achieving Complete Resection
Parameter	Estimatea	OR	95% CI	P Value
ECOG performance status	0.98 vs 0.27	2.65	1.56–4.52	< .001
Residual disease after primary surgery b	0.90 vs 0.27	2.46	1.45–4.20	< .001
Ascites	1.63 vs 0.48	5.08	1.97–13.16	< .001
Localization of recurrence in preoperative diagnostic imaging	0.44 vs 0.31	1.55	0.85–2.82	.155
a See Table 4 for comparison groups. b Alternatively, FIGO stage if residual disease after primary surgery is unknown (HR = 1.87; 95% CI = 1.04–3.37; P = .036). CI = confidence interval; ECOG = Eastern Cooperative Oncology Group; FIGO = International Federation of Gynecology and Obstetrics; HR = hazard ratio; OR = odds ratio. Source: Harter P et al.[28]

The AGO investigators created a predictive model for complete surgical cytoreduction in patients with recurrent disease, which will be investigated prospectively in the AGO-DESKTOP OVAR II clinical trial. In brief, patients with recurrent ovarian cancer who have a disease-free interval > 6 months, an ECOG performance status of 0, no residual disease left after primary surgery, and ascites estimated by diagnostic imaging to be < 500 mL will undergo a laparotomy for surgical cytoreduction, followed by platinum-based chemotherapy. Patients who do not meet all these criteria but for whom surgery is still desired, will undergo an open laparoscopy first to attempt to determine whether a surgical resection is possible. If carcinomatosis is present, they will receive platinum-based chemotherapy. If cytoreductive surgery is possible, they will undergo a laparotomy.

This article reviewed