Approximately 75% of patients with epithelial ovarian cancer present with advanced disease1 and despite aggressive initial treatment, the majority will develop recurrent disease. Even for patients with a negative second-look laparotomy, the cancer may recur in 30–50% of cases.2–4 The standard management of patients with recurrent disease remains poorly defined, particularly the role of surgery.
Secondary cytoreductive surgery is a repeat surgical effort after first-line chemotherapy to further debulk the cancer.5–11 It may be carried out for: 1) progressive disease while undergoing first-line chemotherapy, 2) residual disease discovered at the time of second-look laparotomy, or 3) recurrent cancer after a disease-free interval.
There have been several studies on secondary cytoreductive surgery for patients with recurrent epithelial ovarian cancer.12–17 These studies have included a heterogeneous group of patients, and the results have been difficult to interpret. The introduction of more effective second-line chemotherapeutic agents has given added impetus to the need to define a subset of patients for whom aggressive secondary cytoreduction may be of benefit.
To justify secondary cytoreduction, several concerns need to be clarified, namely: 1) the selection of patients suitable for this approach, 2) the feasibility of achieving optimal residual disease status, 3) the associated morbidity and mortality, and 4) the survival benefit accrued to such patients.
These concerns define the objectives of our retrospective study, which evaluated the outcome of secondary cytoreductive surgery at the Royal Hospital for Women, Sydney, for a group of patients who developed recurrence of an epithelial ovarian cancer after a disease-free interval.
MATERIALS AND METHODS
The study period extended from July 1988 to October 1996, during which 336 patients with invasive epithelial ovarian cancer were treated at the Gynaecological Cancer Centre of the Royal Hospital for Women in Sydney. Their details were prospectively entered into our data bank.
Forty-six patients were identified from our data bank who had undergone secondary cytoreductive surgery for tumor recurrence after a disease-free interval. Most patients were without clinical evidence of disease after their primary operation, but the disease-free interval was calculated from the completion of their primary chemotherapy, provided their tumor markers had returned to normal. All patients with recurrent ovarian cancer were presented to a multidisciplinary Tumor Board, but the decision to operate was ultimately made after a discussion between the gynecologic-oncologist (PTG or NFH) and the patient. Characteristics of their primary disease and treatment are shown in Table 1. Twenty-one patients (46%) had early-stage disease initially.
Preoperative characteristics of the patients are shown in Table 2. Their mean age was 50.3 years, and 23 patients (50%) were symptomatic from their recurrence. The median disease-free interval was 26 months (range 3–136 months) with 41 of 46 patients (89%) having a disease-free interval of at least 12 months.
Laparotomy findings are shown in Table 3. The recurrences were largely confined to the intraperitoneal cavity, and 16 patients (35%) had significant intestinal involvement. Seven patients (15%) had bulky positive lymph nodes.
Univariate survival outcomes were analyzed using the log rank test and survival curves calculated using the method of Kaplan-Meier.18 Multivariable adjustment was performed using proportional hazards regression. Logistic regression was used to examine the effect of prognostic factors on the ability to achieve a complete resection. Data were analyzed using the statistical package ACCorD (Analysis of Censored and Correlated Data, Boffin Software, Sydney, Australia). Survival was calculated from the date of the secondary cytoreductive surgery until the last follow-up or death of the patient.
Nineteen patients (41%) were cytoreduced to no macroscopic disease, and eight patients (18%) had residual disease of more than 2 cm. Resection of involved viscera was required in 20 patients (44%), including intestinal resection in 16 patients (35%), partial liver resection in two patients (4.5%), and partial bladder resection in two patients (4.5%). One patient needed a temporary colostomy, which was subsequently closed without complication.
In two patients, secondary cytoreduction was abandoned after biopsy of the tumor. One had a subdiaphragmatic nodule identified on a computed tomographic scan after a rising CA 125 titer, which appeared to be well localized and resectable. At laparotomy, the tumor mass was found to be densely adherent to the inferior vena cava and the hepatic veins. The other patient also had a localized tumor, but it was densely fixed to the retroperitoneum and extensively involved the base of the small bowel and sigmoid mesentery. Salvage therapy after the secondary cytoreductive surgery varied, with the majority of patients receiving platinum-based chemotherapy (Table 4).
There were six patients who did not receive any postoperative therapy. Three patients had very advanced disease. Two opted for palliative care only, whereas the third patient deteriorated quickly and died of disease 4 weeks after surgery before salvage chemotherapy could be administered.
The remaining three patients had complete resection of solitary tumor recurrences. The CA 125 titers normalized, and the patients were observed without treatment. The CA 125 titers subsequently rose at 9, 12, and 54 months, respectively.
In a multivariable analysis, the ability to achieve complete resection of all macroscopic disease was not influenced by the residual disease after primary debulking surgery, initial International Federation of Gynecology and Obstetrics stage, histologic grade, age, disease-free interval, or the absence of symptoms. However, no patient with disseminated disease had a complete surgical resection, whereas 76% of patients (19 of 25) with localized disease had a complete resection.
Four of 46 patients (8.7%) developed significant postoperative morbidity, which included one case each of severe pulmonary atelectasis requiring temporary artificial ventilation, bronchopneumonia, deep venous thrombosis, and infected vaginal vault hematoma. All of these patients recovered without further surgical intervention. The only mortality in the series was a patient with disseminated disease who developed an intestinal fistula and died 27 days after suboptimal cytoreduction.
The estimated blood loss ranged from 100 mL to 4000 mL, with a mean of 748 mL and a standard deviation of 728 mL. Operating time ranged from 60 to 240 minutes with a mean of 133 minutes (standard deviation 52.5 minutes). The total duration of hospitalization for all patients ranged from 6 to 27 days with a mean of 12 days (standard deviation 4.6 days).
The overall median survival for all patients after secondary cytoreductive surgery was 22.5 months (range 1–142 months). At the end of the study period, nine patients were still alive, three without further disease, at 48, 60, and 142 months, respectively.
Survival by the amount of residual disease present is given in Figure 1. The median survival for patients who had no evidence of macroscopic residual disease (n = 19) was 38 months. By contrast, patients who still had any macroscopic disease (n = 27) had a median survival of only 11 months (P = .002, log rank). Patients with 0–1-cm residual (n = 14) had a median survival of 14.5 months, whereas those with residual greater than 1 cm (n = 13) had a median survival of 11 months.
Figure 2 shows the impact of different disease-free intervals on overall survival. The intervals used were: less than 12 months, 12–24 months, and greater than 24 months. Patients with a longer disease-free interval had a significantly longer median survival. In particular, patients with a disease-free interval of more than 24 months (n = 26) had a median survival of 39 months (P = .001, log rank). The influence of exposure to platinum therapy before secondary cytoreduction was not statistically significant (P = .25, log rank)
The effect of these factors in a multivariable analysis is shown in Figure 3. Patients with no residual disease have a 70% relative risk reduction (95% confidence interval 0.13, 0.71) compared with those with residual disease greater than 1 cm. Patients who have a disease-free interval of greater than 2 years have a 75% relative risk reduction (95% confidence interval 0.08, 0.75) compared with those whose disease-free interval is less than 12 months. Neither initial International Federation of Gynecology and Obstetrics stage (P = .386, log rank) nor tumor grade (P = .52, log rank) influenced survival after secondary cytoreduction.
Cytoreductive surgery has generally been accepted as integral to the primary management of epithelial ovarian cancer,19 but its role in recurrent disease is less well defined. This is one of the largest studies of secondary cytoreductive surgery for recurrent epithelial ovarian cancer.
The patients recruited to previous studies of secondary cytoreductive surgery12–17 have been diverse. Segna et al17 included patients with progressive epithelial ovarian cancer on chemotherapy in their series. Morris et al15 reported 30 patients, and Eisenkop et al12 reported 36 patients with a disease-free interval of more than 6 months. Janicke et al13 reported 30 patients with recurrent disease and a median disease-free interval of 16 months. Our 46 patients had a median disease-free interval of 26 months.
The extent and complexity of surgery has also differed between studies. The prospective study by Eisenkop et al12 employed extensive surgery with the aim of removing all macroscopic disease. Enbloc resection techniques were used including pelvic exenteration and ablative procedures using an argon beam coagulator or cavitron ultrasonic aspirator to eliminate peritoneal or serosal implants. They were able to achieve complete resection of macroscopic disease in 30 patients (83%). In the series reported by Janicke et al,13 where all surgery was planned and carried out by an interdisciplinary team, complete resection was achieved in 47% of patients, and 63% required an intestinal resection. In our series, all surgery was performed by a gynecologic-oncologist, and we were able to resect all macroscopic disease in 41% of patients. Intestinal resection was required in 35%.
Postoperative morbidity and mortality is of concern when surgery of this magnitude is undertaken. The surgical morbidity experienced by Eisenkop et al12 was 30%, including wound infections, septicemia, superior vena caval syndrome, systemic candidiasis, and pseudomembranous colitis. There was one postoperative death (2.8%). Morris et al15 reported 36.7% postoperative morbidity with patients having mainly paralytic ileus. Other morbidity included fever, fungal sepsis, and enterocutaneous fistula. Although there was one incident of intraoperative cardiac arrest, no mortality was reported. Similarly, Janicke et al13 reported 36% of their patients without intestinal anastomosis had prolonged postoperative paralytic ileus, and anastomotic leaks occurred in 16% of patients. The median postoperative stay for patients with intestinal reanastomoses was 26 days and 19 days for those without. In our series, there were five patients (10.8%) with serious postoperative morbidity, one of whom died after development of an enterocutaneous fistula. The occurrence of a postoperative death highlights the need for careful patient selection.
Significant improvement in survival would need to be demonstrated to justify secondary cytoreductive surgery, in view of the usually complex nature of the surgery and the associated serious morbidity and mortality. The presence of any gross residual disease has a negative influence on the patients' survival as demonstrated in previous studies.12–14,16 Our data showed an 11-month median survival for patients with any macroscopic residual disease (n = 27), whereas the median survival for patients without macroscopic residual disease (n = 19) was 38 months.
Our general philosophy was to select patients who had no clinical ascites and in whom the disease was deemed resectable as assessed by clinical examination and radiographic imaging. However, two patients were found to have inoperable disease at laparotomy, and only 41% of patients could have their disease completely resected. It is theoretically possible that initial laparoscopic assessment could help to determine the feasibility of complete resection, but a laparoscopic procedure may be hazardous in a patient with extensive intraperitoneal disease and adhesions from prior laparotomies, and the risk of port-site implantation is another major concern.20
The other factor that significantly affected survival was the disease-free interval. In previous studies,12–17 a better survival outcome has consistently been observed in patients with a longer disease-free interval, although varying cutoff periods of 12, 18, and 24 months have been used. Our data showed a profound difference in the median survival for patients with a disease-free interval of 24 months or more (39 months) compared with those with less than 24 months (11 months). The reasons for this observation probably relate to tumor sensitivity to platinum therapy and intrinsic tumor aggressiveness in terms of rate of growth.
The best salvage therapy after secondary cytoreduction is unknown. In our study, the treatments employed were diverse. Of interest were the three patients who, as a result of the complete resection of their localized recurrent cancer, did not immediately receive any further salvage therapy. These three patients remained in tumor marker remission for 9, 12, and 54 months, respectively. Although they all subsequently developed further tumor recurrence, all returned to their normal lifestyle postoperatively, free from the side effects of further therapy.
About 50% of patients in our study had already been exposed to platinum chemotherapy at the time of their primary treatment. Theoretically, patients who were platinumnaïve could have been expected to have a better survival. However, the difference was not statistically significant.
In conclusion, our data suggest that aggressive secondary cytoreductive surgery can significantly prolong the survival of two groups of patients: those in whom all macroscopic disease can be removed, and those with a disease-free interval of 24 months or longer. The need for resection of bowel and partial resection of other viscera is common, so that morbidity is significant but not prohibitive. Careful patient selection is important, and this type of surgery is applicable to only a small proportion of patients.
1. Petterson F. International Federation of Gynecology and Obstetrics Report. Stockholm, Sweden FIGO, 1988: 110–51.
2. Berek JS, Hacker NF, Lagasse LD, Poth T, Resnick B, Nieberg RK. Second-look laparotomy in Stage III epithelial ovarian cancer: Clinical variables associated with disease status. Obstet Gynecol 1984;64:207–12.
3. Gerhenson DM, Copeland LJ, Wharton JT, Atkinson EN, Sneige N, Edwards CL, et al. Prognosis of surgically determined complete responders in advanced ovarian cancer. Cancer 1985;55:1129–35.
4. Rubin SC, Hoskins WJ, Hakes TB, Markman M, Cain JM, Lewis JL Jr. Recurrence after negative second-look laparotomy for ovarian cancer: Analysis of risk factors. Am J Obstet Gynecol 1988;159:1094–8.
5. Vogl SE, Seltzer V, Calanog AMA, Camacho F, Kaplan BH, Greenwald E. “Second-effort” surgical resection for bulky ovarian cancer. Cancer 1984;54:2220–5.
6. Hoskins WJ, Rubin SC, Dulaney E, Chapman D, Almadrones L, Saigo P, et al. Influence of secondary cytoreduction at the time of second-look laparotomy on the survival of patients with epithelial ovarian carcinoma. Gynecol Oncol 1989;34:365–71.
7. Williams LL. Secondary cytoreduction of ovarian malignancies. In: Markman M, Hoskins WJ, eds. Cancer of the ovary. New York: Raven Press; 1993:187–203.
8. Burke TW, Morris M. Secondary cytoreductive surgery for ovarian cancer. Obstet Gynecol Clin North Amer 1994;21:167–78.
9. Bristow RE, Lagasse LD, Kartan BY. Secondary cytoreduction for advanced epithelial ovarian cancer. Patient selection and review of literature. Cancer 1996;78:2049–62.
10. Hacker NF, van der Burg MEL. Debulking and intervention surgery. Ann Oncol 1993;4(Suppl. 4):S17–S22.
11. Heintz APM. Surgery in ovarian cancer: The concept of cytoreductive surgery. Curr Opin Obstet Gynecol 1996; 8:8–11.
12. Eisenkop SM, Friedman RL, Wang HJ. Secondary cytoreductive surgery for recurrent ovarian cancer. Cancer 1995;76:1606–14.
13. Janicke F, Holscher M, Kuhn W, von Hugo R, Pache L, Siewert JR, et al. Radical surgical procedure improves survival time in patients with ovarian cancer. Cancer 1992;70:2129–36.
14. Vaccarello L, Rubin SC, Vlamis V, Wong G, Jones WB, Lewis JL, et al. Cytoreductive surgery in ovarian carcinoma patients with a documented previously complete surgical response. Gynecol Oncol 1995;57:61–5.
15. Morris M, Gershenson DM, Wharton T, Copeland LJ, Edwards CL, Stringer CA. Secondary cytoreductive surgery for epithelial ovarian cancer. Gynecol Oncol 1989;34:334–8.
16. Berek JS, Hacker NF, Lagasse LD, Nieberg RK, Elashoff RM. Survival of patients following secondary cytoreductive surgery in ovarian cancer. Obstet Gynecol 1983;61:189–93.
17. Segna RA, Dottino PR, Mabdeli JP, Konsker K, Cohen CJ. Secondary cytoreduction for ovarian cancer following cisplatin therapy. J Clin Oncol 1993;11:434–9.
18. Kaplan E, Meier P. Nonparametic estimation from incomplete observations. J Amer Stat Assoc 1958;53:457–81.
19. van der Burg MEL, van Lent M, Buyse M, Kobierska A, Colombo N, Favalli G, et al. The effect of debulking surgery after induction chemotherapy on the prognosis in advanced epithelial ovarian cancer. N Engl J Med 1995; 332:629–34.
20. Childers JM, Aqua KA, Surwit EA, Hallum AV, Hatch KD. Abdominal-wall metastases tumour implantation after laparoscopy for malignant conditions. Obstet Gynecol 1994;84:765–9.