The Significance of Duration of Chemotherapy Interruptions Due to Interval Surgery in Ovarian Cancer Patients Treated With Neoadjuvant Chemotherapy

GYNAECOLOGY GYNAECOLOGY

The Significance of Duration of Chemotherapy Interruptions Due to Interval Surgery in Ovarian Cancer Patients Treated With Neoadjuvant Chemotherapy Tien Le, MD, Kate A. Fathi, BSc (Hon), Laura Hopkins, MD, Wylam Faught, MD, Michael Fung-Kee-Fung, MB Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, University of Ottawa, Ottawa ON

Abstract Objective: To examine the prognostic significance of interruptions to chemotherapy arising from delayed primary surgical debulking following neoadjuvant chemotherapy in women undergoing treatment for ovarian cancer. Methods: We carried out a retrospective chart review to identify women with ovarian cancer who were treated with neoadjuvant chemotherapy and planned delayed primary surgical debulking. Cox regression modelling was used to identify significant predictors of progression-free and overall survival, using well-established prognostic variables and time delays between courses of chemotherapy perioperatively, stratified by residual disease status. Results: Ninety-seven patients with complete data were identified. Their median age was 65.4 years. Fifty-four patients (56%) were left with optimal residual disease (< 1 cm), and 43 patients had suboptimal residual disease. The median delay from the last cycle of chemotherapy to the time of surgery was 29 days (range 20–72). The median delay from surgery to resumption of cytotoxic therapy was 23 days (range 8–65). Chemotherapy courses were interrupted for a median of 50 days (range 36–119) around the time of surgery. No effect was observed on progression-free interval by interruptions to chemotherapy, regardless of residual disease status. With respect to overall survival, the time to resumption of chemotherapy in days and the time delay in days between the two chemotherapy cycles peri-operatively were identified as statistically significant predictors only in patients with suboptimal residual disease. In patients with optimal residual disease status, neither the time of interruption between the two chemotherapy cycles peri-operatively nor the time to resumption of chemotherapy after surgical debulking was significantly predictive of overall survival. Conclusion: In women undergoing treatment for ovarian cancer, the interval between surgery and the resumption of chemotherapy in patients with suboptimal residual disease should be kept as short as possible. Key Words: Neoadjuvant chemotherapy, ovarian cancer, chemotherapy interruption, prognosis Competing Interests: None declared. Received on June 7, 2008 Accepted on July 24, 2008

Résumé Objectif : Examiner l’importance pronostique des interruptions de la chimiothérapie, attribuables à une chirurgie primaire de réduction tumorale différée à la suite d’une chimiothérapie néoadjuvante, chez les femmes subissant un traitement contre le cancer de l’ovaire. Méthodes : Nous avons mené une analyse de dossier rétrospective pour identifier les femmes présentant un cancer de l’ovaire qui ont été traitées au moyen d’une chimiothérapie néoadjuvante et d’une chirurgie primaire de réduction tumorale différée planifiée. Le modèle de régression de Cox a été utilisé pour identifier les prédicteurs importants de survie sans progression et de survie globale, au moyen de variables pronostiques bien établies et des délais entre les cycles périopératoires de chimiothérapie, stratifiés en fonction de l’état quant à la maladie résiduelle. Résultats : Nous avons identifié 97 patientes présentant des données complètes. Leur âge médian était de 65,4 ans. Cinquante-quatre patientes (56 %) présentaient par la suite une maladie résiduelle optimale (< 1 cm), tandis que 43 patientes présentaient une maladie résiduelle sous-optimale. Le délai médian entre le dernier cycle de chimiothérapie et le moment de la chirurgie était de 29 jours (plage : 20-72). Le délai médian entre la chirurgie et la reprise du traitement cytotoxique était de 23 jours (plage : 8-65). Les cycles de chimiothérapie ont été interrompus selon une médiane de 50 jours (plage : 36-119) aux alentours du moment de la chirurgie. Nous n’avons constaté aucun effet attribuable aux interruptions de la chimiothérapie sur l’intervalle sans progression, peu importe l’état quant à la maladie résiduelle. En ce qui concerne la survie globale, le délai en jours jusqu’à la reprise de la chimiothérapie et le délai en jours entre les deux cycles périopératoires de chimiothérapie n’ont été identifiés à titre de prédicteurs significatifs sur le plan statistique que chez les patientes présentant une maladie résiduelle sous-optimale. Chez les patientes présentant un état optimal quant à la maladie résiduelle, ni la durée de l’interruption entre les deux cycles périopératoires de chimiothérapie ni le délai jusqu’à la reprise de la chimiothérapie à la suite de la chirurgie de réduction tumorale n’ont permis de prédire de façon significative la survie globale. Conclusion : Chez les femmes subissant un traitement contre le cancer de l’ovaire qui présentent une maladie résiduelle sous-optimale, l’intervalle entre la chirurgie et la reprise de la chimiothérapie devrait être le plus court possible. J Obstet Gynaecol Can 2009;31(2):161–166

FEBRUARY JOGC FÉVRIER 2009 l

161

GYNAECOLOGY

INTRODUCTION

ost patients with epithelial ovarian cancer present with stage 3 or 4 disease, where extensive extraovarian spread has already occurred.1 Initial aggressive surgical tumour debulking followed by adjuvant combination platinum and taxane-based chemotherapy is often recommended.2 The justification for the initial aggressive surgical approach is based on retrospective and prospective data demonstrating the amount of residual tumour to be strongly predictive of the ultimate prognosis.3–5 It is not uncommon for a patient with advanced ovarian cancer to present in a significantly undernourished state with poor physical performance status and extensive intra-abdominal metastasis. In these patients, radical tumour resection may not be well tolerated, resulting in a high probability of suboptimal debulking. Several recent retrospective reports have shown that patients who undergo a neoadjuvant chemotherapy approach in combination with interval tumour debulking have comparable survival to patients who have standard initial surgical debulking, and they have less post operative morbidity.6–8

M

One theoretical concern in planning major surgical procedures during chemotherapy is the necessary delay between cycles of chemotherapy to allow for hematologic recovery, to minimize postoperative complications, and to optimize wound healing. This interruption in chemotherapy may also allow for continued tumour growth with compromise of outcome. The significance of interrupting chemotherapy in patients treated with neoadjuvant chemotherapy has not been examined in detail. It could be hypothesized that in patients treated in this manner, the tumour bulk remaining after surgery (which further chemotherapy is expected to control) may have greater resistance to further chemotherapy than tumour left behind after primary surgical debulking because of prior exposure to similar cytotoxic agents. It is possible that the interruption of chemotherapy would allow for these chemotherapy-resistant cells to proliferate. In this study, we examined the prognostic impact of an interruption in chemotherapy before and after planned interval surgical debulking following neoadjuvant chemotherapy in women with advanced epithelial ovarian or peritoneal carcinoma. MATERIALS AND METHODS

We carried out a retrospective chart review using the chemotherapy database at the Ottawa General Hospital from the period 1997 to 2002. This study time period was chosen to allow sufficient time for the data to mature. Patients receiving neoadjuvant chemotherapy with documentation 162

l FEBRUARY JOGC FÉVRIER 2009

of planned delayed primary surgical debulking were identified. All debulking surgical procedures were performed by gynaecologic oncologists with the assistance of the resident or fellow on service. Since 1997, all patients seen at our institution with a probable diagnosis of advanced ovarian or peritoneal cancer who have large pelvic-abdominal masses and raised serum CA-125 levels without acute symptoms of bowel obstruction or unmanageable pain are first assessed with a CT scan of the abdomen and pelvis. This is followed by an ultrasound-guided biopsy of the most easily accessible radiographic abnormality. After histologic and immunohistochemical confirmation of adenocarcinoma consistent with a primary ovarian or peritoneal origin, patients are scheduled for treatment with three to four cycles of intravenous carboplatinum (AUC = 5) and paclitaxel (175 mg/m2 over three hours) every three weeks. Delayed primary surgical tumour debulking is scheduled for four weeks after the last chemotherapy cycle. All patients are assessed prior to surgery to ensure that major surgical contraindications are absent and that the management of all concurrent medical comorbidities has been optimized. Following surgery, a similar preoperative chemotherapy regimen is resumed for a further three cycles. The timing of the resumption of chemotherapy is decided jointly by the patient and the attending gynaecologic oncologist, taking into account the patient’s preferences and her postoperative recovery. After completing the prescribed primary treatment (chemotherapy and surgery), patients were followed in the outpatient clinic every three months for four years. At each visit, a complete physical examination including pelvi-rectal examination was performed, with measurement of serum CA125. Suspected disease progression, based on CA125 changes or clinical findings, was confirmed with CT scanning. Subsequent salvage chemotherapy was introduced as appropriate for measurable disease. Standard patient information and disease-related demographics, treatment outcomes, and complications were abstracted manually from medical charts, with crossreferencing to the computerized patient information database at the Ottawa Hospital to corroborate the accuracy of the data. Information on the exact timing of administration of chemotherapy was abstracted from the computerized hospital-based chemotherapy database. Descriptive statistics were used to summarize demographic variables. Cox proportional hazard models were built to identify important factors predictive of progression-free and overall survival, using a conditional backward elimination variable selection strategy to obtain the most parsimonious model. Variables included in the model were age, tumour stage, grade, residual disease status after interval

The Significance of Duration of Chemotherapy Interruptions Due to Interval Surgery in Ovarian Cancer Patients

debulking (optimal residual disease defined as < 1 cm), and either the time interval between two chemotherapy cycles interrupted by planned surgery or the time from surgery to further chemotherapy. Cox models were built for patients with optimal and suboptimal residual disease status separately, because we hypothesized that the effect of delays in chemotherapy might be different in these two groups. Kaplan Meier survival analysis was used to estimate the survival functions. Log rank tests were used to compare survival curves. All P values less than 0.05 were considered to be statistically significant. All data analysis was performed using SPSS version 15 for Windows (SPSS Inc., Chicago IL). The study protocol was approved by the University of Ottawa Research Ethics Board. RESULTS

Over the five-year study period, 101 patients were treated with neoadjuvant chemotherapy. This made up about 80% of all ovarian cancers seen and treated during this period, as the neoadjuvant approach was our preferred treatment option. Ninety-seven patients with available complete data were included in the study cohort. The median age of the group was 65.4 years. Seventy-four percent of the patients had no associated significant medical comorbidity, with the remainder having their medical conditions well controlled at time of initial diagnosis. The most commonly noted medical comorbidity was hypertension, which was present in 16% of all patients. Most patients presented with stage III or IV disease (93%) and grade 3 tumours (73%). The demographic distribution with respect to stage of tumour, grade, and final histology based on findings at time of surgical debulking is summarized in Table 1. All patients received combination neoadjuvant chemotherapy with intravenous carboplatinum (AUC = 5) and paclitaxel (175mg/m2) over three hours. No dose delays or reduction because of excessive toxicity was necessary during the neoadjuvant treatment period. A complete response after neoadjuvant chemotherapy (based on clinical examination findings and CA125 criteria) was observed in 19 patients (19.6%) and a partial response in 58 patients (59.8%). There was no progression of disease during neoadjuvant chemotherapy. All patients proceeded to have debulking surgery irrespective of their response to neoadjuvant treatment. The median time delay from the last cycle of chemotherapy to surgery was 29 days (range 20–72). Following surgery, 26 patients (27%) had all palpable tumour removed, leaving only microscopic residual disease. Overall, 54 patients (56%) were left with optimal residual disease (< 1 cm), and 43 patients had suboptimal residual disease. There was no significant difference in demographics or the type of surgical procedure between

Table 1. Distribution of disease characteristics (N = 97) Demographic characteristics

Patients, n (%)

Tumour stage 2

7 (7.2)

3

79 (81.4)

4

11 (11.4)

Tumour grade 1

4 (4.1)

2

22 (22.7)

3

71 (73.2)

Final tumour histology Serous

82 (84.5)

Mucinous

1 (1.0)

Endometroid

9 (9.3)

Clear cell

3 (3.1)

Mixed histologies

2 (2.1)

those with optimal and suboptimal residual disease after surgery. Radical upper abdominal tumour debulking procedures were not commonly used during this time period. The median delay from surgery to resumption of chemotherapy was 23 days (range 8–65). Chemotherapy was interrupted for a median of 50 days (range 36–119) around the time of debulking surgery. At the conclusion of the primary chemotherapy protocol after surgical debulking, 51.5% of patients achieved a complete response and 45.4% had a partial response based on clinical and CA125 criteria. At a median follow up of 32 months (range 6.2 to 107), 21 patients (22%) had died following disease progression. The median survival for the group was 79.3 months (95% CI 68.4- 90.2). Cox regression models revealed that in patients with optimal residual disease status, lower disease stage was the only significant predictor of a prolonged progression-free survival. In patients with suboptimal residual disease, older age was the only significant predictor of a prolonged progression-free survival. Neither time delays between the two chemotherapy cycles peri-operatively nor the time to resumption of chemotherapy after surgical debulking was significantly predictive of progression free survival in all patients. With respect to overall survival, the time to chemotherapy resumption in days (HR 1.06; 95% CI 1.01–1.12, P = 0.021) and the delay in days between the two chemotherapy cycles perioperatively (HR 1.05; 95% CI 1.01–1.08, P = 0.007) were identified as statistically significant predictors in those patients who had suboptimal residual disease. The estimated hazard ratios, corresponding 95% confidence intervals, and P values for the chemotherapy time delay variable FEBRUARY JOGC FÉVRIER 2009 l

163

GYNAECOLOGY

Table 2. Effects of chemotherapy interruptions on overall survivals stratified by residual disease status (Cox Models) Variable

HR

95% CI

P

Time between chemotherapy cycles (suboptimal residuals)

1.05

1.01–1.08

0.01

Time to resumption of chemotherapy after surgery (suboptimal residuals)

1.06

1.01–1.12

0.02

Time between chemotherapy cycles (optimal residuals)

0.99

0.97–1.08

0.86

Time to resumption of chemotherapy after surgery (optimal residuals)

1.01

0.97–1.08

0.68

HR: hazard ratio

in the Cox models for patients with optimal and suboptimal residual disease status are summarized in Table 2. In patients with optimal residual status, neither the time interruptions between the two chemotherapy cycles peri-operatively nor the time to resumption of chemotherapy after surgical debulking was significantly predictive of overall survival. DISCUSSION

In our study, the time delay between peri-operative cycles of chemotherapy and the delay in resumption of chemotherapy after surgery significantly influenced overall survival time only for the subset of patients with suboptimal residual disease. This might have resulted from the spontaneous development of a chemotherapy resistant clonogenic cell population in large tumour residuals that were not surgically removed, and the waning controlling effects of prior chemotherapy exposure in these large tumour masses with poor chemotherapy penetration. The evolution of new multiple drug resistant (MDR) clones would not be expected to influence the time to first clinical progression, as documented in this study, because of their relatively small number shortly after surgery. However, MDR clones can negatively influence the response to further salvage chemotherapy, thus shortening patients’ overall survival. Cytoreductive surgery has been proven to be effective as the primary treatment for advanced stage ovarian carcinoma; the extent of tumour residual is a strong prognostic factor in predicting progression free and overall survival. A recent meta-analysis of 81 cohorts of stage III and IV ovarian cancer patients estimated that for each 10% increase in maximal surgical cytoreduction achieved there was a corresponding 5.5% increase in median survival time.9 Theoretical benefits of surgical debulking include the removal of large tumour masses with poor blood supply containing cells that are in the Gompertzian growth cycle with slow 164

l FEBRUARY JOGC FÉVRIER 2009

mitotic rates. Debulking surgery would ideally leave behind smaller tumour masses with a higher mitotic rate10,11 and optimize sensitivity to chemotherapy.12 In addition, the risk of spontaneous evolution of chemotherapy resistant clones would be decreased.13 Furthermore, the removal of large intra-abdominal metastases should allow bowel function to return to normal with subsequent improved nutritional status and well-being for the patient. Gynaecologic oncologists are becoming increasingly aggressive in tumour debulking efforts by employing radical pelvic and upper abdominal debulking procedures, because this has been documented to benefit survival rates.14,15 This change in practice can be expected to result in an increase in peri-operative morbidity. As the cornerstone for the management of advanced ovarian cancer remains the combination of surgery and adjuvant cytotoxic chemotherapy, surgical morbidity has the potential to delay the resumption of chemotherapy. This may partially negate the expected benefits of the surgery because of tumour regrowth. Significant perioperative morbidity has been shown to be an adverse prognostic factor in the initial management of ovarian cancer.16 It is unclear for how long the postoperative start of adjuvant chemotherapy can be delayed, to allow patients to recover from surgery, without compromising their prognosis. Flynn et al.17 studied 472 ovarian cancer patients enrolled in four clinical trials by the Scottish Gynaecological Cancer Trials Group, all of whom had received platinum-containing chemotherapy (either in combination with a taxane or cyclophosphamide), to determine whether the interval from primary surgery to the beginning of chemotherapy had an effect on progression-free survival. Patients were stratified according to when their adjuvant chemotherapy began, using the median time from surgery to chemotherapy of 22 days as established in the study database. The analysis was stratified by study and arm or cohort within study to remove any possible influence of the different studies and study drug doses. Multivariate analysis was

The Significance of Duration of Chemotherapy Interruptions Due to Interval Surgery in Ovarian Cancer Patients

then performed to adjust for disease stage, residual disease, and patients’ performance status. Univariate analysis of the above median and below median groups demonstrated poorer progression-free survival for those with earlier chemotherapy treatment (HR 0.84; 95% CI 0.67–1.06, P = 0.14). However, those patients treated earlier had bulkier residual disease (> 2 cm; P = 0.006). When multivariate analysis was performed incorporating residual disease status, FIGO stage, and performance status, the hazard ratio for interval to surgery was 0.99 (95% CI 0.79–1.24, P = 0.91). Similarly, Gadducci et al.18 studied the effect of time from primary surgery to the start of taxane plus platinum-based chemotherapy on the prognosis of patients with advanced ovarian cancer. The 25%, 50%, and 75% quartiles of time intervals from surgery to the start of chemotherapy were 11, 21, and 31 days, respectively. These values were used to stratify the study cohort. In this study, only the amount of residual disease (P = 0.001) and stage (IIc to III vs. IV; P = 0.04) were independent prognostic variables for survival. No significant differences in complete response rates and survival times were observed between patients with an interval from surgery to chemotherapy of less than 11 days, 12 to 21 days, 22 to 31 days, or longer than 31 days. About two thirds of patients in this study underwent second-look surgery after finishing their first-line treatment, and one half of these patients were found to have persistent disease. All of these patients and some of the pathologic complete responders immediately received further chemotherapy, making comparison of overall time to progression and survival difficult to assess. The findings in these two large retrospective studies indicate that a delay in starting adjuvant chemotherapy after primary surgical debulking of between 21 and 31 days does not seem to have a significant adverse prognostic effect on patient survival. Tumour cell proliferation kinetics, however, may be different after neoadjuvant chemotherapy exposure. Several animal studies have demonstrated accelerated tumour repopulation after chemotherapy, leading to a requirement for increased doses with each subsequent course of chemotherapy for tumour control.19–21 Such repopulation might also limit the effectiveness of chemotherapy, accounting for the poor response to treatment.22 Reassuringly, our results did not indicate that the interruption between chemotherapy cycles because of planned surgery or the time delays in resumption of chemotherapy after surgery were significant predictors of time to first clinical progression, regardless of residual disease status. This could be explained by the fact that most patients in our study population had a clinical response to the neoadjuvant platinum based regimen, and the cytology of the residual disease in these patients was more homogenous and relatively more resistant to

platinum. As a result, the expected response to further platinum-based chemotherapy would be less than to the initial therapy, diluting the potential benefit of immediate resumption of chemotherapy. Furthermore, it can be hypothesized that these tumours are still affected by prior chemotherapy cycles, preventing the expected accelerated cell repopulation after surgical debulking. This hypothesis is consistent with the finding by Davis et al.23 in a pilot study of patients treated with platinum-based chemotherapy for ovarian cancer. In this study, at variable times after the last chemotherapy (mean 33 days) the proliferative index (assessed by staining for a nuclear antigen associated with proliferation, Ki67) was increased in only four patients, reduced in 12 and unchanged in five patients. Similarly, Fisher et al.12 assessed how the variation in the time interval between primary tumour removal and cyclophosphamide administration affected residual tumour cell kinetics and animal survival in a murine mammary adenocarcinoma. These authors noted that when cyclophosphamide was given before the operation, it completely prevented the increase in labelling index resulting from tumour removal, and prolonged survival to the greatest extent. Our study was limited by its retrospective design, with unavoidable selection bias and potential for unadjusted confounding variables such as patients’ performance status or a need for reduction in chemotherapy doses. Because of the relatively small number of patients, we could not incorporate all prognostic indicators in the multivariate model. Our findings will need to be prospectively confirmed in a much larger dataset. Although our estimated hazard ratios for the duration of chemotherapy interruptions in suboptimally debulked patients peri-operatively were statistically significant, the lower limits of the 95% confidence interval for each estimate were very close to 1, and their clinical significance must be viewed with caution in counselling patients. As was shown in our previous publication,24 chemotherapy can be administered quickly in the postoperative period with no obvious increase in immediate postoperative morbidity or impaired wound healing. As residual disease status is recognized as the most important prognostic factor that can be directly influenced by the surgical oncologist, radical surgical efforts are justified to achieve optimal residual disease, and concerns about the possible delay in chemotherapy resumption because of anticipated higher postoperative morbidity should not reduce these efforts. Furthermore, patients who feel they are not ready to resume chemotherapy immediately after optimal surgery can be reassured that their prognosis will not be adversely affected. FEBRUARY JOGC FÉVRIER 2009 l

165

GYNAECOLOGY

CONCLUSION

In patients having suboptimal debulked disease at interval surgery, all efforts should be made to minimize the duration of the interval between interval surgery and resumption of chemotherapy. However, in patients having optimal debulking of disease, a reasonable delay in the resumption of chemotherapy has not been shown to have an adverse effect on survival. REFERENCES 1. Holschneider CH, Berek JS. Ovarian cancer: epidemiology, biology, and prognostic factors. Semin Surg Oncol 2000;19(1):3–10. 2. Berkenblit A, Cannistra SA. Advances in the management of epithelial ovarian cancer. J Reprod Med 2005;50(6):426–38. 3. Omura GA, Brady MF, Homesley HD, Yordan E, Major FJ, Buchsbaum HJ, et al. Long-term follow-up and prognostic factor analysis in advanced ovarian carcinoma: the Gynecologic Oncology Group experience. J Clin Oncol 1991;9(7):1138–50. 4. Van Houwelingen JC, ten Bokkel Huinink WW, van der Burg ME, van Oosterom AT, Neijt JP, et al. Predictability of the survival of patients with advanced ovarian cancer. J Clin Oncol 1989;7(6):769–73. 5. Hoskins WJ. Surgical staging and cytoreductive surgery of epithelial ovarian cancer. Cancer 1993;71(4 Suppl):1534–40. 6. Morice P, Brehier-Ollive D, Rey A, Atallah D, Lhomme C, Pautier P, et al. Results of interval debulking surgery in advanced stage ovarian cancer: an exposed-non-exposed study. Ann Oncol 2003 Jan;14(1):74–7. 7. Vergote I, De Wever I, Tjalma W, Van Gramberen M, Decloedt J, van Dam P. Neoadjuvant chemotherapy or primary debulking surgery in advanced ovarian carcinoma: a retrospective analysis of 285 patients. Gynecol Oncol 1998 Dec;71(3):431–6. 8. Lawton FG, Redman CW, Luesley DM, Chan KK, Blackledge G. Neoadjuvant (cytoreductive) chemotherapy combined with intervention debulking surgery in advanced, unresected epithelial ovarian cancer. Obstet Gynecol 1989 Jan;73(1):61–5. 9. Bristow RE, Tomacruz RS, Armstrong DK, Trimble EL, Montz FJ. Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. J Clin Oncol 2002;20(5):1248–59. 10. Gunduz N, Fisher B, Saffer EA. Effect of surgical removal on the growth and kinetics of residual tumor. Cancer Res 1979;39(10):3861–5. 11. Simpson-Herren L, Sanford AH, Holmquist JP. Effects of surgery on the cell kinetics of residual tumor. Cancer Treat Rep 1976;60(12):1749–60.

166

l FEBRUARY JOGC FÉVRIER 2009

12. Fisher B, Gunduz N, Saffer EA. Influence of the interval between primary tumor removal and chemotherapy on kinetics and growth of metastases. Cancer Res 1983;43(4):1488–92. 13. Goldie JH, Coldman AJ. The genetic origin of drug resistance in neoplasms: implications for systemic therapy. Cancer Res 1984;44(9):3643–53. 14. Chi DS, Franklin CC, Levine DA, Akselrod F, Sabbatini P, Jarnagin WR, et al. Improved optimal cytoreduction rates for stages IIIC and IV epithelial ovarian, fallopian tube, and primary peritoneal cancer: a change in surgical approach. Gynecol Oncol 2004;94(3):650–4. 15. Eisenhauer EL, Abu-Rustum NR, Sonoda Y, Levine DA, Poynor EA, Aghajanian C, et al. The addition of extensive upper abdominal surgery to achieve optimal cytoreduction improves survival in patients with stages IIIC-IV epithelial ovarian cancer. Gynecol Oncol 2006;103(3):1083–90. 16. Le T, Alshaikh G, Hopkins L, Faught W, Fung MF. Prognostic significance of postoperative morbidities in patients with advanced epithelial ovarian cancer treated with neoadjuvant chemotherapy and delayed primary surgical debulking. Ann Surg Oncol 2006;13(12):1711–6. 17. Flynn PM, Paul J, Cruickshank DJ; Scottish Gynaecological Cancer Trials Group. Does the interval from primary surgery to chemotherapy influence progression-free survival in ovarian cancer? Gynecol Oncol 2002;86(3):354–7. 18. Gadducci A, Sartori E, Landoni F, Zola P, Maggino T, Maggioni A, et al. Relationship between time interval from primary surgery to the start of taxane- plus platinum-based chemotherapy and clinical outcome of patients with advanced epithelial ovarian cancer: results of a multicenter retrospective Italian study. J Clin Oncol 2005;23(4):751–8. 19. Stephens TC, Peacock JH. Tumour volume response, initial cell kill and cellular repopulation in B16 melanoma treated with cyclophosphamide and 1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea. Br J Cancer 1977;36:313–21. 20. Milas L, Nakayama T, Hunter N, Jones S, Lin TM, Yamada S, et al. Dynamics of tumour cell clonogen repopulation in a murine sarcoma treated with cyclophosphamide. Radiother Oncol 1994;30:247–53. 21. Wu L, Tannock IF. Repopulation in murine breast tumours during and after sequential treatments with cyclophosphamide and 5-fluorouracil. Cancer Res 2003;63:2134–8. 22. Bourhis J, Wilson G, Wibault P, Janot F, Bosq J, Armand JP, et al. Rapid tumour cell proliferation after induction chemotherapy in oropharyngeal cancer. Laryngoscope 1994;104:468–72. 23. Davis AJ, Chapman W, Hedley DW, Oza AM, Tannock IF. Assessment of tumour cell repopulation after chemotherapy for advanced ovarian cancer: pilot study. Cytometry 2003;A51:1–6. 24. Le T, Faught W, Hopkins L, Fung Kee Fung M. Primary chemotherapy and adjuvant tumor debulking in the management of advanced-stage epithelial ovarian cancer. Int J Gynecol Cancer 2005;15(5):770–5.