- Email: [email protected]
Interval debulking surgery in advanced epithelial ovarian cancer
Best Practice & Research Clinical Obstetrics and Gynaecology Vol. 16, No. 4, pp. 573±583, 2002
doi:10.1053/beog.2002.0302, available online at http://www.idealibrary.com on
8 Interval debulking surgery in advanced epithelial ovarian cancer Sergio Pecorelli
MD, PhD
Franco Odicino
MD
Department of Obstetrics & Gynaecology, Division of Gynaecological Oncology, University of Brescia, Spedali Civili di Brescia, Italy and European Institute of Oncology, Milan, Italy
Department of Obstetrics and Gynaecology, University of Brescia, Italy
Giuseppe Favalli
MD
European Institute of Oncology, Milan, Italy
Cytoreductive surgery and chemotherapy are the mainstay for the treatment of advanced epithelial ovarian cancer. In order to minimize the tumour burden before chemotherapy, cytoreductive surgery is usually performed ®rst. The importance of the amount of residual disease as the main prognostic factor for patients suering from advanced disease has been almost universally accepted even in the absence of prospective randomized trials addressing the bene®t of cytoreductive surgery. In the last decade, the value of debulking surgery after induction chemotherapy ± interval debulking surgery, IDS ± has been widely debated, especially after the completion of a prospective randomized study from the EORTC addressing the introduction of a surgical procedure with debulking intent preceded and followed by cytoreductive chemotherapy. The rationale of such a strategy in the context of the primary treatment of advanced ovarian cancer lies in a higher cytoreductibility to the `optimal' status forwarded, and possibly facilitated, by chemotherapy. The results demonstrated a prolongation of both progression-free survival and median survival in favour of patients randomized to IDS (5 and 6 months, respectively). Multivariate analysis revealed IDS to be an independent prognostic factor which reduced the risk of death by 33% at 3 years and by 48% in subsequent re-evaluation after more than 6 years of observation. Despite the above, results have been questioned by many, leading the GOG to perform a similar study which has been concluded very recently. Nevertheless, the main concern regarding the application of IDS in all instances relates to the morbidity of two major surgical procedures integrated within a short period during which cytotoxic chemotherapy is also administered. Neoadjuvant chemotherapy has been recently proposed to avoid a non-useful surgical procedure in patients considered `optimally unresectable' after diagnosis of advanced ovarian cancer. Whether or not this newer approach will translate into a longer survival with a better quality of life is going to be addressed by a novel EORTC study. Finally, the concept of a `chemical' cytoreduction preceding and facilitating a subsequent `surgical' eort has been recently introduced also in the treatment of recurrent disease. The EORTC has recently initiated a prospective randomized study (LOROCSON ± Late Onset Recurrent Ovarian Cancer: Address for correspondence: 2a Divisione di Ginecologia e Ostetricia, Spedali Civili di Brescia, Piazzale Spedali Civili, 1, I-25123, Italy. c 2002 Published by Elsevier Science Ltd. 1521±6934/02/$ - see front matter *
574 S. Pecorelli, F. Odicino and G. Favalli
Surgery or Not) to validate the importance of such an approach to be balanced with medical treatment alone not only in terms of survival but also as far as quality of life is concerned. Key words: ovarian neoplasms; surgery; combined modality therapy; re-operation.
Appropriate surgery remains the cornerstone of management for patients aected by advanced epithelial ovarian cancer. Surgery is required for adequate staging and for cytoreduction of advanced bulky disease. Cytoreduction is de®ned as maximum tumour debulking in order to improve the eectiveness of further chemotherapy. Virtually all studies of advanced ovarian cancer indicate that the amount of residual disease at the completion of the primary surgery is the most important prognostic indicator. The hypothetical rationale for debulking surgery is an increase in chemosensitivity of the residual tumour lesions. With the removal of the relatively insensitive large tumour masses with poor central blood supply and low growth fractions, the sensitivity to cytotoxic drugs is increased by a better perfusion of the small residual tumour lesions and a higher growth fraction. Moreover, the chance of developing primary drug-resistant clones by spontaneous mutations, as well as the probability of developing drug-induced resistance, is lower in small tumour foci. Finally, a signi®cant decrease in tumour volume improves the general conditions of the patient, resulting in a higher qualityof life, possibly in an enhanced immunocompetence and, thus, a better outcome. Since the mid 1970s Griths1 reported that the duration of survival was directly related to the diameter of the largest residual disease remaining after cytoreductive surgery. Thereafter, many studies con®rmed that the diameter of the largest residual disease was directly related to the response rate2, the progression-free interval3, the likelihood of being negative at second surgical evaluation4, and the survival.5 These observations led to the concept of `optimal residual disease' as the ultimate goal of primary debulking, with tumour residual foci less than 2 or 1 cm in largest diameter. The problem still under debate is whether the observed improved prognosis is a function of surgical skill, of the tumour biology, or of both. The rationale of the bene®ts obtained with cytoreductive surgery is pertinent to the increased sensitivity of tumour cells to chemotherapy in the presence of small residual foci. Chemosensitivity may therefore be related to the size of the residual tumour nodules according to the rules of the Gompertzian model for solid tumours. However, considering the re-growth of tumour cells occurring between cycles of chemotherapy, the theoretical bene®t of cytoreductive surgery should be clinically valuable only in patients with residual disease less than 1 g.6 Therefore these concepts have to be taken into consideration in de®ning the `optimal residual disease' diameter, which has been changed during the last few decades from 2 cm to the `no residual' disease status of the present time. While the goal of an `optimal residual disease' when de®ned as `tumour nodules less than 1±2 cm' is achievable in almost 70% of cases7, the possibility of achieving the true `optimal residual disease' de®ned as `no residual disease status after primary cytoreduction' is only 25% of all advanced epithelial ovarian cancer patients.8 Therefore, in recent years new strategies for a better integration of chemotherapy with surgery have been developed, performed and validated by means of clinical trials. INTERVAL DEBULKING SURGERY Interval debulking (or cytoreductive) surgery (IDS) can be de®ned as `a surgical procedure with debulking intent preceded and followed by chemotherapy' during
Interval debulking surgery 575
primary treatment of advanced epithelial ovarian cancer. The value of such a strategy (surgical procedure after induction chemotherapy) has been debated over the past few decades. Several studies supported the rationale for introducing IDS in the context of the primary treatment of advanced epithelial ovarian cancer, even if the ®nal clinical impact on prognosis remained to be further investigated. Moreover, in the late 1980s data concerning such an approach were contradictory. Some studies indicated that patients in whom cytoreduction achieved an optimal status after induction chemotherapy had approximately the same survival rate as did patients in whom the same goal could be achieved at primary surgery.9,10 Other authors, however, reported the opposite.11±13 In these latter series patients with optimal cytoreduction at IDS had, indeed, poorer survival rates compared with patients who had been optimally cytoreduced at primary surgery. Moreover, the survival of patients with optimal cytoreduction at IDS was the same as in patients with suboptimal cytoreduction. However, all of these studies included only small numbers of patients bearing dierent prognostic factors and treated with dierent and sometimes experimental chemotherapy regimens. Irrespective of the above, the de®nitive answer on the potential bene®ts of interval cytoreductive surgery awaited the completion of prospective randomized studies. In April 1986 the West Midlands Group started a prospective randomized study to investigate whether intervention debulking surgery could improve survival in patients with advanced ovarian cancer with bulky (42 cm) residual disease after primary surgery. Unfortunately the study enrolled only 86 patients of whom 79 were available for the ®nal analysis.14 Despite an increased cytoreductive activity of the combined modality approach (73% of the patients submitted to IDS achieved an optimal ± cut-o at 2 cm ± residual disease) no signi®cant improvement in survival could be demonstrated in comparison with the standard approach (primary cytoreductive surgery followed by a full course of chemotherapy). Nonetheless, this study, published in 1994, con®rmed that such an approach was clinically feasible, with little severe morbidity, and led to an estimated reduction in mortality of 0.71 (95% CI: 0.44±1.33) for the IDS arm, which, albeit not statistically signi®cant due to the low statistical power of the study, deserved further investigation in larger series. In March 1987 the Gynaecological Cancer Cooperative Group (GCCG) of the EORTC (European Organisation for Research and Treatment of Cancer) initiated a randomized prospective multicentre clinical trial to investigate the value of debulking surgery after induction chemotherapy in advanced ovarian cancer patients with residual disease after primary surgery of more than 1 cm. All patients received an induction chemotherapy of three courses of cisplatin-cyclophosphamide (75 mg/m2 and 750 mg/m2, respectively) after primary surgery. Patients not progressing after the third course were randomly assigned to debulking surgery or to no surgery after strati®cation for performance status, response to chemotherapy and centre. All randomized patients received at least another three cycles of cisplatin-cyclophosphamide regimen. The end points of the study were overall survival and progressionfree survival. Some 425 patients were enrolled in the study from 1987 to 1993. Treatment data were available from 278 of the 319 randomized patients when the trial was published in 199515; 140 patients had been allocated to intervention surgery (7% did not undergo surgery) and 138 to no surgery (none had surgery). Patients' characteristics and prognostic factors were well balanced between the two treatment arms. The administered chemotherapy and treatment delays were identical in both groups, indicating that IDS did not interfere with the post-operative chemotherapy. Surgical
576 S. Pecorelli, F. Odicino and G. Favalli
data were available from 127 patients. At IDS laparotomy, 44 patients had lesions of less than 1 cm as a result of induction chemotherapy. In 83 patients (65%) the lesions measured more than 1 cm. In 37 of these patients the lesions could be surgically reduced to less than 1 cm, and in 17 patients to diameters of 1±2 cm. Lesions of more than 2 cm had to be left behind in 29 patients (22.8%). Data on the possibility of cytoreducing after induction chemotherapy at IDS are reported in Table 1. Surgery was not associated with death or severe morbidity. Intra-operative complications were observed in 5% of the patients and mild post-operative complications in 14%. At the ®rst analysis the median follow-up for the evaluable patients was 3.5 years. Both the survival and progression-free survival were lengthened (P 5 0.01) for the patients assigned the interval debulking surgery. The median survival was 26 months for the IDS patients and 20 months for the no-surgery patients, while the median progression-free survival was 18 months and 13 months respectively. Interestingly, the survival and progression-free survival patterns of the patients with lesions less than 1 cm before and after cytoreduction (IDS) were signi®cantly better compared to those of the patients with a suboptimal cytoreduction at primary surgery. Nonetheless, the survival and progression-free survival patterns of the patients with suboptimal debulking surgery did not signi®cantly dier from those of patients in the no-surgery arm. In the multivariate analysis, IDS was an independent prognostic factor for survival (P 0.012). Overall, after adjustment for all prognostic factors, IDS reduced the risk of death by 33% (P 0.008) (Table 2). In a further analysis performed in 199816, when all prognostic factors were simultaneously taken into account in a Cox regression model, the reduction in risk of death increased to 60% with a median follow-up of 6.3 years. In Figures 1 and 2 both the survival and progression-free survival are shown as reported in the 1998 analysis. Ultimately, even when the patients were subdivided in high- and low-risk groups, based on either individual prognostic factors or the multivariate risk, there was no dierence in the relative bene®t of survival between the two groups (high and low risk). Therefore, it was not possible, even after the most recent update, to identify a subgroup of patients who did not bene®t from IDS. It was concluded that the increases of 5 and 6 months in median progression-free and overall survival outweigh the low morbidity associated with surgery and that IDS should be considered in patients with suboptimal primary surgery even if it cannot replace the primary debulking surgery which remains the gold standard in the management of advanced epithelial ovarian carcinoma. Despite these well known reported data, the conclusions of the EORTC-GCCG interval debulking study have been questioned since its publication. Questions
Table 1. Results of debulking surgery after induction chemotherapy. Redman et al 199414
van der Burg et al 199515
IDS not performed Microscopic disease after IDS
12/37 (32.4%) 5/26 (19.2%)
Optimal residual disease after IDS Suboptimal residual disease after IDS
14/26 (53.8%) 7/26a (27%)
10/140 (7.1%) 48/127 (37.8%) (3 patients not evaluated) 33/127 (26%) 46/127b (36.2%)
a
Optimal residual disease 4 2 cm. Optimal residual disease 5 1 cm.
b
Interval debulking surgery 577 Table 2. Analysis of the risk of death after interval debulking surgerya. Risk reduction (%) Unadjusted Univariate Number of lesions Ascites Tumour grade Response after 3 induction CT WHO Perf. status FIGO stage Multivariate (Cox model)
P value
31
0.012
36 35 34 33 32 29 33
0.003 0.004 0.006 0.006 0.011 0.026 0.008
aReproduced
from van den Burg MEL et al (1995, New England Journal of Medicine 332: 629±634) with permission.
are related mainly to the issue of quality of life and to the long-term follow-up results.8,17±19 While concerns over long-term follow-up results have been answered by the analysis performed in 1998, where the observed bene®t in IDS patients seemed not only to last but also eventually to improve20, the issue of the quality of life is still open. While proponents of debulking surgery claim that it improves the patient's quality of life, opponents assert the opposite. Unfortunately there are no published studies addressing this issue in a prospective manner. The only paper speci®cally addressing the quality of life issue with respect to debulking surgery and ovarian cancer concluded that, despite brief survival, patients optimally debulked experienced a `good' quality of life which was slightly better than that of those patients suboptimally debulked.21
Survival by treatment 100 90 80 70 60 50 40 30 20 10
p = 0.0032
0
2
4 years
6
8
10
O
N
122
159
84
40
16
5
Surgery
138
160
64
21
10
4
No surgery
Number of patients at risk:
Treatment
Figure 1. Interval debulking surgery versus no surgery: overall survival. Reproduced from van der Burg MEL et al (1997, International Journal of Gynecological Cancer 7:) with permission.
578 S. Pecorelli, F. Odicino and G. Favalli
100 90 80 70 60 50 40 30 20 10
Progression Free Survival by treatment p=0.0055
2
0
4
6
8
10
years O
N
Patients at risk:
Treatment
138
150
45
22
9
4
Surgery
148
160
30
10
4
2
No surgery
Figure 2. Interval debulking surgery versus no surgery: progression free survival. Reproduced from van der Burg MEL et al (1997, International Journal of Gynecological Cancer 7:) with permission.
After more than 6 years since its publication, we believe that the comments made by Hacker in his editorial written in 199622, underlining the importance of the EORTC study, still remain valid: `this study does not imply that interval cytoreduction is the treatment of choice, but rather it gives added validity to the concept of primary cytoreduction, which should still be regarded as the gold standard for most patients. Clearly not all patients will derive equal bene®t from aggressive cytoreduction, but to deny it to all is to deny it to those for whom it will undoubtedly prolong survival'. That is because, as reported above, in the attempt to identify those patients not bene®ting from interval debulking surgery, van der Burg and co-workers have performed a thorough multivariate analysis and have found that the bene®t of surgery remained signi®cant even after adjustment for all other known prognostic factors. Thus, a subgroup of patients who clearly did not bene®t from IDS or a subgroup of patients in whom the bene®t was much greater than the overall bene®t could not be identi®ed. This fact, once more, underlined the need for further investigations. In this respect, given the clinical impact of the conclusions provided by the EORTC-GCCG trial, a con®rmatory prospective randomized study was activated in June 1994 by the Gynecologic Oncology Group (GOG protocol 152) in which the chemotherapy regimen consisted of the newer association cisplatin (75 mg/m2) and paclitaxel (175 mg/m2) and the protocol designed as in the European study. This trial has been recently closed after accruing more than 500 patients (425 patients randomized up to January 2001) and results are awaited with great interest in the near future. COMMENTS The European experience on the value of a second surgical procedure (IDS) within the primary chemo-surgical treatment for advanced epithelial ovarian cancer has led to important developments not only as far as the primary treatment is concerned but also in planning the treatment of recurrent disease.
Interval debulking surgery 579
In searching for a better-integrated chemo-surgical treatment for a patient suering from advanced epithelial ovarian cancer, one should, indeed, be concerned about the number of surgical procedures to be performed in order to reach a possible longer survival. A `primary laparotomy procedure' (with diagnostic and possibly debulking intent) followed by an `interval debulking surgery' (with optimal debulking intent after three courses of chemotherapy) and, ®nally ± in selected instances ± a `third-look surgical procedure' (either laparotomic or laparoscopic) might have an adverse impact on patients' quality of life. For this reason, but not only for it, in more recent years neoadjuvant chemotherapy has been proposed for patients with established bulky disease (stage IIIc).23,24 Neoadjuvant chemotherapy is de®ned as a `chemotherapy administered before any surgical debulking procedure'. The concept of such an `induction' cytoreductive chemotherapy has been derived from the positive experience of the combined chemo and surgical cytoreduction of the EORTC-GCCG interval debulking trial, and from retrospective studies investigating the issue of chemical cytoreduction preceding a more eective surgical cytoreduction. In one retrospective study the survival of a group of patients with advanced ovarian cancer treated with chemotherapy when unfavourable characteristics were present was similar to a former series of `optimally debulked' (at ®rst laparotomy) patients treated at the same Institution. An optimal cytoreduction was obtained in 89% of patients submitted to neoadjuvant chemotherapy, suggesting that the same survival, with lower operative morbidity, could be obtained with an initial chemical debulking compared with a surgical one.25 More recently, a French multicentre study has reported the experience of six gynaecological oncology departments where 91% optimal surgical cytoreduction was achieved in 54 patients treated by neoadjuvant chemotherapy after being considered as `optimally unresectable' at primary surgical exploration (laparotomic or laparoscopic).26 It can be concluded, along with the French authors, that `neoadjuvant chemotherapy for primary unresectable ovarian cancer leads to the selection of a subset of patients sensitive to chemotherapy in whom optimal cytoreduction can be achieved after chemotherapy by standard surgery in a high proportion of cases. Conversely, aggressive surgery can be avoided in patients with initial chemo resistance in whom the prognosis is known to be poor regardless of treatment'. However, the major problem with the studies reported above is the nonrandomized nature of the comparison groups and the small size of the neoadjuvant chemotherapy patient population leading to the inability to detect clinically important dierences. The EORTC-GCCG will be randomizing approximately 800 women with advanced ovarian cancer to conventional upfront surgery followed by platinum-taxane-based chemotherapy or to platinum-taxane-based neoadjuvant chemotherapy followed by surgery (EORTC protocol 55971); this study will hopefully provide a de®nitive conclusion for such an important issue. The concept of `chemical' cytoreduction preceding and facilitating a subsequent `surgical' cytoreduction has recently been introduced in the treatment strategy for recurrent epithelial ovarian cancer. Indeed, despite the most recent advances in the integrated chemo-surgical management of advanced ovarian cancer, most patients will eventually recur and die of their disease. Although the best salvage treatment of recurrent ovarian cancer has not yet been established, a re-challenge with platinumcontaining regimens seems to be the treatment of choice for recurrent platinumsensitive patients after a platinum-free interval longer than 6±12 months.27 Even in this more favourable category of patients, after a response rate to platinum re-challenging
580 S. Pecorelli, F. Odicino and G. Favalli
regimens of about 40±90%, the median survival is only 18 months.28,29 Patients with recurrent pelvic and abdominal tumours are only occasionally candidates for a surgical excision of their disease. This operation is generally de®ned as `secondary cytoreductive surgery' and tumour resection should be restricted to carefully selected patients for whom resection has a reasonable chance of either prolonging life or signi®cantly palliating symptoms. The role that surgery might play in the treatment of recurrent ovarian cancer has been addressed in few retrospective studies, including a limited series of cases, all showing that those patients in whom the residual disease could be completely resected survived de®nitely longer than those in whom it could not.30±34 In these series of patients with recurrences of advanced epithelial ovarian cancer, an optimal debulking was achieved in 60±80% of cases, leading to a median survival of 18 months even if the surgical procedure was never performed after an `induction chemotherapy', according to the most recent concepts emerging from the IDS experience in primary treatment. A thorough analysis of these series led to the conclusions that the ideal candidates for a surgical cytoreduction for recurrent ovarian cancer are those suering from a late recurrence (more than 1 year from the conclusion of primary treatment), solitary rather than multiple, pelvic rather than abdominal or retroperitoneal and, especially, when reasonable evidence exists that no residual disease will be left behind after the debulking attempt. So far, as suggested by Pecorelli at the NIH Consensus Conference on Ovarian Cancer in 199426, `when the principle of interval debulking surgery is applied to the patient with recurrent disease, only patients responding to chemotherapy would undergo surgery during the course of therapy, with the hope of achieving optimal cytoreduction. If these ®ndings are proven signi®cant by a well planned randomized trial (chemotherapy integrated with interval debulking surgery versus chemotherapy alone) we could not only achieve a better survival, or a similar survival with a better quality of life, but we could also spare these patients unnecessary surgery'. In an attempt to answer this issue in a prospective and randomized fashion, the EORTC started the LOROCSON trial (Late Onset Recurrent Ovarian Cancer: Surgery or Not ± GCCG protocol 55963) in 2000 in which `secondary surgery' follows a chemotherapy induction (three courses of platinum-based treatment before and after the surgical cytoreduction) and is compared to an exclusive platinum-based re-challenging of the patient with late (more than 1 year) recurrent disease. In this trial, quality of life issues (addressed throughout the QLQ-c30 core questionnaire of the EORTC) represent an important issue to be balanced with survival and progression-free survival of the recurrent disease.
CONCLUSIONS Despite the fact that during the past two decades the cytoreductive attitude of gynaecological oncologists has increased, and a higher response rate to platinumtaxane chemotherapy has been achieved, the state-of-the-art treatment still fails to cure the vast majority of patients suering from advanced epithelial ovarian cancer. After many debates concerning the optimal integration of surgery and chemotherapy, we can undoubtedly re-arm the strong correlation between chemosensitivity, successful debulking surgery and survival from this disease. This correlation strongly supports the concept that it is the biological characteristics of the tumour that allow the patient to have successful cytoreductive surgery rather than the aggressiveness of surgery (surgeon's skill and attitude) itself.
Interval debulking surgery 581
Interval debulking surgery after primary sub-optimal cytoreductive surgery followed by induction platinum-based chemotherapy has been proven, at least according to the EORTC experience, to be eective in prolonging both progressionfree and overall survival for advanced epithelial ovarian cancer. A con®rmatory trial from the GOG is expected to validate the European results. The main concern related to interval cytoreductive surgery after a short course of chemotherapy following the primary laparotomy procedure is about the morbidity of two major surgical procedures during a short period of time in which cytotoxic chemotherapy is also administered. Neoadjuvant chemotherapy has been therefore recently proposed for patients with bulky disease with the theoretical advantage of improvement of the patient's quality of life prior to and after the operative procedure. Whether or not the supposed increased rate of optimal cytoreduction will translate into an improvement in survival or progression-free interval is still to be de®ned. Finally, the value of a secondary cytoreduction for patients recurrences after a long therapy-free interval is still under debate and needs to be balanced with medical treatment alone not only in relation to survival but with respect to a thorough evaluation of the quality of life. Practice points . proposal to include in the clinical management of advanced epithelial ovarian cancer patients a new integrated chemotherapeutic and surgical approach for those patients who cannot bene®t from primary cytoreductive surgery in achieving an `optimal' residual disease status . explain the background of the rationale for debulking surgery . include the data of those studies supporting the rationale to introduce interval debulking surgery in the context of the primary treatment of advanced epithelial ovarian cancer patients . include data of prospective clinical trials on interval debulking surgery
Research agenda . the value of interval debulking surgery in the management of advanced epithelial ovarian cancer patients who are not supposed to achieve an `optimal' residual disease status from primary cytoreductive surgery
REFERENCES 1. Grith CTH. Surgical resection of tumor bulk in the primary treatment of ovarian carcinoma. National Cancer Institute Monographs 1975; 42: 1010±1014. 2. Hoskins WJ et al. The eect of the diameter of largest residual disease on survival after primary cytoreductive surgery in patients with suboptimal residual epithelial ovarian carcinoma. American Journal of Obstetrics & Gynecology 1994; 170: 974±980. 3. Piver MS et al. The impact of aggressive debulking surgery and cisplatin based chemotherapy on progression-free survival in stage III and IV ovarian carcinoma. Journal of Clinical Oncology 1988; 6: 983±989.
582 S. Pecorelli, F. Odicino and G. Favalli 4. Heintz APM et al. The treatment of advanced ovarian carcinoma: clinical variables associated with prognosis. Gynecologic Oncology 1988; 30: 347±358. 5. Hacker NF et al. Primary cytoreductive surgery for epithelial ovarian cancer. Obstetrics & Gynecology 1983; 61: 413±420. 6. De Vita VT. Principles of chemotherapy. In De Vita VT et al (eds) Cancer: Principles and Practice of Oncology, 4th edn, pp 997±. Philadelphia, PA: Lippincott-Raven, 1997. 7. Grith CT et al. Role of cytoreductive surgical therapy in the management of advanced ovarian cancer. Cancer Treatment Report 1979; 63: 335±340. 8. Covens AL. A critique of surgical cytoreduction in advanced ovarian cancer. Gynecologic Oncology 2000; 78: 269±274. 9. Nejit JP, tenBokkel Huinink WW, van der Burg MEL et al. Randomized trial comparing two combination chemotherapy regimens (CHAP-5 v CP) in advanced ovarian carcinoma. Journal of Clinical Oncology 1987; 5: 1157±1168. 10. Jacob JH, Gersherson DM, Morris M et al. Neoadjuvant chemotherapy and interval debulking for advanced epithelial ovarian cancer. Gynecologic Oncology 1991; 42: 146±150. 11. Nejit JP, tenBokkel Huinink WW, van der Burg MEL et al. Long-term survival in ovarian cancer: mature data from the Netherland Joint Study Group for Ovarian Cancer. European Journal of Cancer 1991; 27: 1367±1372. 12. Nejit JP, tenBokkel Huinink WW, van den Burg MEL et al. Randomized trial comparing the combination chemotherapy regimens (CHAP-5 vs CP) in advanced ovarian carcinoma. Journal of Clinical Oncology 1987; 5: 1157±1168. 13. Varma R, Redman CWE, Blackledge G et al. Early second surgery in bulky epithelial ovarian cancer does not improve survival: results of a randomized study. Abstract of the 2nd meeting of IGCS, Toronto, October 9±13, 1989. 14. Redman CWE, Warwick J, Luesley DM et al. Intervention debulking surgery in advanced epithelial ovarian cancer. British Journal of Obstetrics and Gynaecology 1994; 101: 142±146. 15. van der Burg MEL, van Lent M, Buyse M et al. The eect of debulking surgery after induction chemotherapy on the prognosis in advanced epithelial ovarian cancer. New England Journal of Medicine 1995; 332: 629±634. 16. van der Burg MEL et al. After 6 years follow-up intervention debulking surgery remains an independent prognostic factor for survival in advanced epithelial ovarian cancer: an EORTC-GCCG study. International Journal of Gynecological Cancer 1997; 7: . 17. Kehoe S, Sha® M & Luesley D. Interval cytoreduction in ovarian cancer (letter). New England Journal of Medicine 1995; 333: 254. 18. Neijt JP. Interval cytoreduction in ovarian cancer (letter). New England Journal of Medicine 1995; 333: 254±255. 19. Cannistra SA. Interval cytoreduction in ovarian cancer (letter). New England Journal of Medicine 1995; 333: 255. 20. Van der Burg MEL, Buyse M & Pecorelli S. Interval cytoreduction in ovarian cancer (letter). New England Journal of Medicine 1995; 333: 255. 21. Blythe JG & Wahl TP. Debulking surgery: does it increase the quality of survival? Gynecologic Oncology 1982; 14: 396±408. 22. Hacker NF. Cytoreduction for advanced ovarian cancer in perspective. International Journal of Gynecological Cancer 1996; 6: 159±160. 23. Lawton FG et al. Neoadjuvant `cytoreductive' chemotherapy combined with intervention debulking surgery in advanced, unresected epithelial ovarian cancer. Obstetrics & Gynecolology 1989; 73: 61±65. 24. Surwit E et al. Neoadjuvant chemotherapy for advanced ovarian cancer. International Journal of Gynecological Cancer 1996; 6: 356±361. 25. Vergote I, De Wever I, Tjalma W et al. Neoadjuvant chemotherapy or primary debulking surgery in advanced ovarian carcinoma: a retrospective analysis of 285 patients. Gynecological Oncology 1998; 71: 431±436. 26. Ansquer Y, Leblanc E, Clough K et al. Neoadjuvant chemotherapy for unresectable ovarian carcinoma. Cancer 2001; 91: 2329±2334. 27. National Institute of Health Issue Consensus Development Conference Statement: Ovarian Cancer. Screening Treatment and Follow up. Gynecologic Oncology 1993; 4: 83±88. 28. Gore ME, Fryatt B, Wiltshaw & Dawson T. Treatment of relapsed carcinoma of the ovary with cisplatin or carboplatin following initial treatment with these compounds. Gynecologic Oncology 1990; 36: 207±211. 29. Markman M, Rothman R, Hakes T et al. Second line platinum therapy in patients with ovarian cancer previously treated with cisplatin. Journal of Clinical Oncology 1991; 9: 383±393. 30. Janicke F, Holscher M, Kuhn W et al. Radical surgical procedure improves survival time in patients with recurrent ovarian cancer. Cancer 1992; 70: 2129±2136.
Interval debulking surgery 583 31. Munkarah AR, Gershenson DM, Burke TE et al. Secondary cytoreductive surgery for a clinically apparent solitary intra-abdominal recurrence in patients with epithelial ovarian cancer. Proceedings of the American Society of Clinical Oncology 1996; 15: 295 (abstract). 32. Segna RA, Dottino PR, Mandeli JP et al. Secondary cytoreduction for ovarian cancer following cisplatin therapy. Journal of Clinical Oncology 1993; 11: 434±439. 33. Morris M, Gershenson DM, Taylor Wharton J et al. Secondary cytoreductive surgery for recurrent epithelial ovarian cancer. Gynecologic Oncology 1989; 34: 334±338. 34. Pecorelli S, Sartori E & Santin AD. Follow-up after primary therapy: management of the symptomatic patient. Surgery Gynecologic Oncology 1994; 55S: 138±142.
doi:10.1053/beog.2002.0302, available online at http://www.idealibrary.com on
8 Interval debulking surgery in advanced epithelial ovarian cancer Sergio Pecorelli
MD, PhD
Franco Odicino
MD
Department of Obstetrics & Gynaecology, Division of Gynaecological Oncology, University of Brescia, Spedali Civili di Brescia, Italy and European Institute of Oncology, Milan, Italy
Department of Obstetrics and Gynaecology, University of Brescia, Italy
Giuseppe Favalli
MD
European Institute of Oncology, Milan, Italy
Cytoreductive surgery and chemotherapy are the mainstay for the treatment of advanced epithelial ovarian cancer. In order to minimize the tumour burden before chemotherapy, cytoreductive surgery is usually performed ®rst. The importance of the amount of residual disease as the main prognostic factor for patients suering from advanced disease has been almost universally accepted even in the absence of prospective randomized trials addressing the bene®t of cytoreductive surgery. In the last decade, the value of debulking surgery after induction chemotherapy ± interval debulking surgery, IDS ± has been widely debated, especially after the completion of a prospective randomized study from the EORTC addressing the introduction of a surgical procedure with debulking intent preceded and followed by cytoreductive chemotherapy. The rationale of such a strategy in the context of the primary treatment of advanced ovarian cancer lies in a higher cytoreductibility to the `optimal' status forwarded, and possibly facilitated, by chemotherapy. The results demonstrated a prolongation of both progression-free survival and median survival in favour of patients randomized to IDS (5 and 6 months, respectively). Multivariate analysis revealed IDS to be an independent prognostic factor which reduced the risk of death by 33% at 3 years and by 48% in subsequent re-evaluation after more than 6 years of observation. Despite the above, results have been questioned by many, leading the GOG to perform a similar study which has been concluded very recently. Nevertheless, the main concern regarding the application of IDS in all instances relates to the morbidity of two major surgical procedures integrated within a short period during which cytotoxic chemotherapy is also administered. Neoadjuvant chemotherapy has been recently proposed to avoid a non-useful surgical procedure in patients considered `optimally unresectable' after diagnosis of advanced ovarian cancer. Whether or not this newer approach will translate into a longer survival with a better quality of life is going to be addressed by a novel EORTC study. Finally, the concept of a `chemical' cytoreduction preceding and facilitating a subsequent `surgical' eort has been recently introduced also in the treatment of recurrent disease. The EORTC has recently initiated a prospective randomized study (LOROCSON ± Late Onset Recurrent Ovarian Cancer: Address for correspondence: 2a Divisione di Ginecologia e Ostetricia, Spedali Civili di Brescia, Piazzale Spedali Civili, 1, I-25123, Italy. c 2002 Published by Elsevier Science Ltd. 1521±6934/02/$ - see front matter *
574 S. Pecorelli, F. Odicino and G. Favalli
Surgery or Not) to validate the importance of such an approach to be balanced with medical treatment alone not only in terms of survival but also as far as quality of life is concerned. Key words: ovarian neoplasms; surgery; combined modality therapy; re-operation.
Appropriate surgery remains the cornerstone of management for patients aected by advanced epithelial ovarian cancer. Surgery is required for adequate staging and for cytoreduction of advanced bulky disease. Cytoreduction is de®ned as maximum tumour debulking in order to improve the eectiveness of further chemotherapy. Virtually all studies of advanced ovarian cancer indicate that the amount of residual disease at the completion of the primary surgery is the most important prognostic indicator. The hypothetical rationale for debulking surgery is an increase in chemosensitivity of the residual tumour lesions. With the removal of the relatively insensitive large tumour masses with poor central blood supply and low growth fractions, the sensitivity to cytotoxic drugs is increased by a better perfusion of the small residual tumour lesions and a higher growth fraction. Moreover, the chance of developing primary drug-resistant clones by spontaneous mutations, as well as the probability of developing drug-induced resistance, is lower in small tumour foci. Finally, a signi®cant decrease in tumour volume improves the general conditions of the patient, resulting in a higher qualityof life, possibly in an enhanced immunocompetence and, thus, a better outcome. Since the mid 1970s Griths1 reported that the duration of survival was directly related to the diameter of the largest residual disease remaining after cytoreductive surgery. Thereafter, many studies con®rmed that the diameter of the largest residual disease was directly related to the response rate2, the progression-free interval3, the likelihood of being negative at second surgical evaluation4, and the survival.5 These observations led to the concept of `optimal residual disease' as the ultimate goal of primary debulking, with tumour residual foci less than 2 or 1 cm in largest diameter. The problem still under debate is whether the observed improved prognosis is a function of surgical skill, of the tumour biology, or of both. The rationale of the bene®ts obtained with cytoreductive surgery is pertinent to the increased sensitivity of tumour cells to chemotherapy in the presence of small residual foci. Chemosensitivity may therefore be related to the size of the residual tumour nodules according to the rules of the Gompertzian model for solid tumours. However, considering the re-growth of tumour cells occurring between cycles of chemotherapy, the theoretical bene®t of cytoreductive surgery should be clinically valuable only in patients with residual disease less than 1 g.6 Therefore these concepts have to be taken into consideration in de®ning the `optimal residual disease' diameter, which has been changed during the last few decades from 2 cm to the `no residual' disease status of the present time. While the goal of an `optimal residual disease' when de®ned as `tumour nodules less than 1±2 cm' is achievable in almost 70% of cases7, the possibility of achieving the true `optimal residual disease' de®ned as `no residual disease status after primary cytoreduction' is only 25% of all advanced epithelial ovarian cancer patients.8 Therefore, in recent years new strategies for a better integration of chemotherapy with surgery have been developed, performed and validated by means of clinical trials. INTERVAL DEBULKING SURGERY Interval debulking (or cytoreductive) surgery (IDS) can be de®ned as `a surgical procedure with debulking intent preceded and followed by chemotherapy' during
Interval debulking surgery 575
primary treatment of advanced epithelial ovarian cancer. The value of such a strategy (surgical procedure after induction chemotherapy) has been debated over the past few decades. Several studies supported the rationale for introducing IDS in the context of the primary treatment of advanced epithelial ovarian cancer, even if the ®nal clinical impact on prognosis remained to be further investigated. Moreover, in the late 1980s data concerning such an approach were contradictory. Some studies indicated that patients in whom cytoreduction achieved an optimal status after induction chemotherapy had approximately the same survival rate as did patients in whom the same goal could be achieved at primary surgery.9,10 Other authors, however, reported the opposite.11±13 In these latter series patients with optimal cytoreduction at IDS had, indeed, poorer survival rates compared with patients who had been optimally cytoreduced at primary surgery. Moreover, the survival of patients with optimal cytoreduction at IDS was the same as in patients with suboptimal cytoreduction. However, all of these studies included only small numbers of patients bearing dierent prognostic factors and treated with dierent and sometimes experimental chemotherapy regimens. Irrespective of the above, the de®nitive answer on the potential bene®ts of interval cytoreductive surgery awaited the completion of prospective randomized studies. In April 1986 the West Midlands Group started a prospective randomized study to investigate whether intervention debulking surgery could improve survival in patients with advanced ovarian cancer with bulky (42 cm) residual disease after primary surgery. Unfortunately the study enrolled only 86 patients of whom 79 were available for the ®nal analysis.14 Despite an increased cytoreductive activity of the combined modality approach (73% of the patients submitted to IDS achieved an optimal ± cut-o at 2 cm ± residual disease) no signi®cant improvement in survival could be demonstrated in comparison with the standard approach (primary cytoreductive surgery followed by a full course of chemotherapy). Nonetheless, this study, published in 1994, con®rmed that such an approach was clinically feasible, with little severe morbidity, and led to an estimated reduction in mortality of 0.71 (95% CI: 0.44±1.33) for the IDS arm, which, albeit not statistically signi®cant due to the low statistical power of the study, deserved further investigation in larger series. In March 1987 the Gynaecological Cancer Cooperative Group (GCCG) of the EORTC (European Organisation for Research and Treatment of Cancer) initiated a randomized prospective multicentre clinical trial to investigate the value of debulking surgery after induction chemotherapy in advanced ovarian cancer patients with residual disease after primary surgery of more than 1 cm. All patients received an induction chemotherapy of three courses of cisplatin-cyclophosphamide (75 mg/m2 and 750 mg/m2, respectively) after primary surgery. Patients not progressing after the third course were randomly assigned to debulking surgery or to no surgery after strati®cation for performance status, response to chemotherapy and centre. All randomized patients received at least another three cycles of cisplatin-cyclophosphamide regimen. The end points of the study were overall survival and progressionfree survival. Some 425 patients were enrolled in the study from 1987 to 1993. Treatment data were available from 278 of the 319 randomized patients when the trial was published in 199515; 140 patients had been allocated to intervention surgery (7% did not undergo surgery) and 138 to no surgery (none had surgery). Patients' characteristics and prognostic factors were well balanced between the two treatment arms. The administered chemotherapy and treatment delays were identical in both groups, indicating that IDS did not interfere with the post-operative chemotherapy. Surgical
576 S. Pecorelli, F. Odicino and G. Favalli
data were available from 127 patients. At IDS laparotomy, 44 patients had lesions of less than 1 cm as a result of induction chemotherapy. In 83 patients (65%) the lesions measured more than 1 cm. In 37 of these patients the lesions could be surgically reduced to less than 1 cm, and in 17 patients to diameters of 1±2 cm. Lesions of more than 2 cm had to be left behind in 29 patients (22.8%). Data on the possibility of cytoreducing after induction chemotherapy at IDS are reported in Table 1. Surgery was not associated with death or severe morbidity. Intra-operative complications were observed in 5% of the patients and mild post-operative complications in 14%. At the ®rst analysis the median follow-up for the evaluable patients was 3.5 years. Both the survival and progression-free survival were lengthened (P 5 0.01) for the patients assigned the interval debulking surgery. The median survival was 26 months for the IDS patients and 20 months for the no-surgery patients, while the median progression-free survival was 18 months and 13 months respectively. Interestingly, the survival and progression-free survival patterns of the patients with lesions less than 1 cm before and after cytoreduction (IDS) were signi®cantly better compared to those of the patients with a suboptimal cytoreduction at primary surgery. Nonetheless, the survival and progression-free survival patterns of the patients with suboptimal debulking surgery did not signi®cantly dier from those of patients in the no-surgery arm. In the multivariate analysis, IDS was an independent prognostic factor for survival (P 0.012). Overall, after adjustment for all prognostic factors, IDS reduced the risk of death by 33% (P 0.008) (Table 2). In a further analysis performed in 199816, when all prognostic factors were simultaneously taken into account in a Cox regression model, the reduction in risk of death increased to 60% with a median follow-up of 6.3 years. In Figures 1 and 2 both the survival and progression-free survival are shown as reported in the 1998 analysis. Ultimately, even when the patients were subdivided in high- and low-risk groups, based on either individual prognostic factors or the multivariate risk, there was no dierence in the relative bene®t of survival between the two groups (high and low risk). Therefore, it was not possible, even after the most recent update, to identify a subgroup of patients who did not bene®t from IDS. It was concluded that the increases of 5 and 6 months in median progression-free and overall survival outweigh the low morbidity associated with surgery and that IDS should be considered in patients with suboptimal primary surgery even if it cannot replace the primary debulking surgery which remains the gold standard in the management of advanced epithelial ovarian carcinoma. Despite these well known reported data, the conclusions of the EORTC-GCCG interval debulking study have been questioned since its publication. Questions
Table 1. Results of debulking surgery after induction chemotherapy. Redman et al 199414
van der Burg et al 199515
IDS not performed Microscopic disease after IDS
12/37 (32.4%) 5/26 (19.2%)
Optimal residual disease after IDS Suboptimal residual disease after IDS
14/26 (53.8%) 7/26a (27%)
10/140 (7.1%) 48/127 (37.8%) (3 patients not evaluated) 33/127 (26%) 46/127b (36.2%)
a
Optimal residual disease 4 2 cm. Optimal residual disease 5 1 cm.
b
Interval debulking surgery 577 Table 2. Analysis of the risk of death after interval debulking surgerya. Risk reduction (%) Unadjusted Univariate Number of lesions Ascites Tumour grade Response after 3 induction CT WHO Perf. status FIGO stage Multivariate (Cox model)
P value
31
0.012
36 35 34 33 32 29 33
0.003 0.004 0.006 0.006 0.011 0.026 0.008
aReproduced
from van den Burg MEL et al (1995, New England Journal of Medicine 332: 629±634) with permission.
are related mainly to the issue of quality of life and to the long-term follow-up results.8,17±19 While concerns over long-term follow-up results have been answered by the analysis performed in 1998, where the observed bene®t in IDS patients seemed not only to last but also eventually to improve20, the issue of the quality of life is still open. While proponents of debulking surgery claim that it improves the patient's quality of life, opponents assert the opposite. Unfortunately there are no published studies addressing this issue in a prospective manner. The only paper speci®cally addressing the quality of life issue with respect to debulking surgery and ovarian cancer concluded that, despite brief survival, patients optimally debulked experienced a `good' quality of life which was slightly better than that of those patients suboptimally debulked.21
Survival by treatment 100 90 80 70 60 50 40 30 20 10
p = 0.0032
0
2
4 years
6
8
10
O
N
122
159
84
40
16
5
Surgery
138
160
64
21
10
4
No surgery
Number of patients at risk:
Treatment
Figure 1. Interval debulking surgery versus no surgery: overall survival. Reproduced from van der Burg MEL et al (1997, International Journal of Gynecological Cancer 7:) with permission.
578 S. Pecorelli, F. Odicino and G. Favalli
100 90 80 70 60 50 40 30 20 10
Progression Free Survival by treatment p=0.0055
2
0
4
6
8
10
years O
N
Patients at risk:
Treatment
138
150
45
22
9
4
Surgery
148
160
30
10
4
2
No surgery
Figure 2. Interval debulking surgery versus no surgery: progression free survival. Reproduced from van der Burg MEL et al (1997, International Journal of Gynecological Cancer 7:) with permission.
After more than 6 years since its publication, we believe that the comments made by Hacker in his editorial written in 199622, underlining the importance of the EORTC study, still remain valid: `this study does not imply that interval cytoreduction is the treatment of choice, but rather it gives added validity to the concept of primary cytoreduction, which should still be regarded as the gold standard for most patients. Clearly not all patients will derive equal bene®t from aggressive cytoreduction, but to deny it to all is to deny it to those for whom it will undoubtedly prolong survival'. That is because, as reported above, in the attempt to identify those patients not bene®ting from interval debulking surgery, van der Burg and co-workers have performed a thorough multivariate analysis and have found that the bene®t of surgery remained signi®cant even after adjustment for all other known prognostic factors. Thus, a subgroup of patients who clearly did not bene®t from IDS or a subgroup of patients in whom the bene®t was much greater than the overall bene®t could not be identi®ed. This fact, once more, underlined the need for further investigations. In this respect, given the clinical impact of the conclusions provided by the EORTC-GCCG trial, a con®rmatory prospective randomized study was activated in June 1994 by the Gynecologic Oncology Group (GOG protocol 152) in which the chemotherapy regimen consisted of the newer association cisplatin (75 mg/m2) and paclitaxel (175 mg/m2) and the protocol designed as in the European study. This trial has been recently closed after accruing more than 500 patients (425 patients randomized up to January 2001) and results are awaited with great interest in the near future. COMMENTS The European experience on the value of a second surgical procedure (IDS) within the primary chemo-surgical treatment for advanced epithelial ovarian cancer has led to important developments not only as far as the primary treatment is concerned but also in planning the treatment of recurrent disease.
Interval debulking surgery 579
In searching for a better-integrated chemo-surgical treatment for a patient suering from advanced epithelial ovarian cancer, one should, indeed, be concerned about the number of surgical procedures to be performed in order to reach a possible longer survival. A `primary laparotomy procedure' (with diagnostic and possibly debulking intent) followed by an `interval debulking surgery' (with optimal debulking intent after three courses of chemotherapy) and, ®nally ± in selected instances ± a `third-look surgical procedure' (either laparotomic or laparoscopic) might have an adverse impact on patients' quality of life. For this reason, but not only for it, in more recent years neoadjuvant chemotherapy has been proposed for patients with established bulky disease (stage IIIc).23,24 Neoadjuvant chemotherapy is de®ned as a `chemotherapy administered before any surgical debulking procedure'. The concept of such an `induction' cytoreductive chemotherapy has been derived from the positive experience of the combined chemo and surgical cytoreduction of the EORTC-GCCG interval debulking trial, and from retrospective studies investigating the issue of chemical cytoreduction preceding a more eective surgical cytoreduction. In one retrospective study the survival of a group of patients with advanced ovarian cancer treated with chemotherapy when unfavourable characteristics were present was similar to a former series of `optimally debulked' (at ®rst laparotomy) patients treated at the same Institution. An optimal cytoreduction was obtained in 89% of patients submitted to neoadjuvant chemotherapy, suggesting that the same survival, with lower operative morbidity, could be obtained with an initial chemical debulking compared with a surgical one.25 More recently, a French multicentre study has reported the experience of six gynaecological oncology departments where 91% optimal surgical cytoreduction was achieved in 54 patients treated by neoadjuvant chemotherapy after being considered as `optimally unresectable' at primary surgical exploration (laparotomic or laparoscopic).26 It can be concluded, along with the French authors, that `neoadjuvant chemotherapy for primary unresectable ovarian cancer leads to the selection of a subset of patients sensitive to chemotherapy in whom optimal cytoreduction can be achieved after chemotherapy by standard surgery in a high proportion of cases. Conversely, aggressive surgery can be avoided in patients with initial chemo resistance in whom the prognosis is known to be poor regardless of treatment'. However, the major problem with the studies reported above is the nonrandomized nature of the comparison groups and the small size of the neoadjuvant chemotherapy patient population leading to the inability to detect clinically important dierences. The EORTC-GCCG will be randomizing approximately 800 women with advanced ovarian cancer to conventional upfront surgery followed by platinum-taxane-based chemotherapy or to platinum-taxane-based neoadjuvant chemotherapy followed by surgery (EORTC protocol 55971); this study will hopefully provide a de®nitive conclusion for such an important issue. The concept of `chemical' cytoreduction preceding and facilitating a subsequent `surgical' cytoreduction has recently been introduced in the treatment strategy for recurrent epithelial ovarian cancer. Indeed, despite the most recent advances in the integrated chemo-surgical management of advanced ovarian cancer, most patients will eventually recur and die of their disease. Although the best salvage treatment of recurrent ovarian cancer has not yet been established, a re-challenge with platinumcontaining regimens seems to be the treatment of choice for recurrent platinumsensitive patients after a platinum-free interval longer than 6±12 months.27 Even in this more favourable category of patients, after a response rate to platinum re-challenging
580 S. Pecorelli, F. Odicino and G. Favalli
regimens of about 40±90%, the median survival is only 18 months.28,29 Patients with recurrent pelvic and abdominal tumours are only occasionally candidates for a surgical excision of their disease. This operation is generally de®ned as `secondary cytoreductive surgery' and tumour resection should be restricted to carefully selected patients for whom resection has a reasonable chance of either prolonging life or signi®cantly palliating symptoms. The role that surgery might play in the treatment of recurrent ovarian cancer has been addressed in few retrospective studies, including a limited series of cases, all showing that those patients in whom the residual disease could be completely resected survived de®nitely longer than those in whom it could not.30±34 In these series of patients with recurrences of advanced epithelial ovarian cancer, an optimal debulking was achieved in 60±80% of cases, leading to a median survival of 18 months even if the surgical procedure was never performed after an `induction chemotherapy', according to the most recent concepts emerging from the IDS experience in primary treatment. A thorough analysis of these series led to the conclusions that the ideal candidates for a surgical cytoreduction for recurrent ovarian cancer are those suering from a late recurrence (more than 1 year from the conclusion of primary treatment), solitary rather than multiple, pelvic rather than abdominal or retroperitoneal and, especially, when reasonable evidence exists that no residual disease will be left behind after the debulking attempt. So far, as suggested by Pecorelli at the NIH Consensus Conference on Ovarian Cancer in 199426, `when the principle of interval debulking surgery is applied to the patient with recurrent disease, only patients responding to chemotherapy would undergo surgery during the course of therapy, with the hope of achieving optimal cytoreduction. If these ®ndings are proven signi®cant by a well planned randomized trial (chemotherapy integrated with interval debulking surgery versus chemotherapy alone) we could not only achieve a better survival, or a similar survival with a better quality of life, but we could also spare these patients unnecessary surgery'. In an attempt to answer this issue in a prospective and randomized fashion, the EORTC started the LOROCSON trial (Late Onset Recurrent Ovarian Cancer: Surgery or Not ± GCCG protocol 55963) in 2000 in which `secondary surgery' follows a chemotherapy induction (three courses of platinum-based treatment before and after the surgical cytoreduction) and is compared to an exclusive platinum-based re-challenging of the patient with late (more than 1 year) recurrent disease. In this trial, quality of life issues (addressed throughout the QLQ-c30 core questionnaire of the EORTC) represent an important issue to be balanced with survival and progression-free survival of the recurrent disease.
CONCLUSIONS Despite the fact that during the past two decades the cytoreductive attitude of gynaecological oncologists has increased, and a higher response rate to platinumtaxane chemotherapy has been achieved, the state-of-the-art treatment still fails to cure the vast majority of patients suering from advanced epithelial ovarian cancer. After many debates concerning the optimal integration of surgery and chemotherapy, we can undoubtedly re-arm the strong correlation between chemosensitivity, successful debulking surgery and survival from this disease. This correlation strongly supports the concept that it is the biological characteristics of the tumour that allow the patient to have successful cytoreductive surgery rather than the aggressiveness of surgery (surgeon's skill and attitude) itself.
Interval debulking surgery 581
Interval debulking surgery after primary sub-optimal cytoreductive surgery followed by induction platinum-based chemotherapy has been proven, at least according to the EORTC experience, to be eective in prolonging both progressionfree and overall survival for advanced epithelial ovarian cancer. A con®rmatory trial from the GOG is expected to validate the European results. The main concern related to interval cytoreductive surgery after a short course of chemotherapy following the primary laparotomy procedure is about the morbidity of two major surgical procedures during a short period of time in which cytotoxic chemotherapy is also administered. Neoadjuvant chemotherapy has been therefore recently proposed for patients with bulky disease with the theoretical advantage of improvement of the patient's quality of life prior to and after the operative procedure. Whether or not the supposed increased rate of optimal cytoreduction will translate into an improvement in survival or progression-free interval is still to be de®ned. Finally, the value of a secondary cytoreduction for patients recurrences after a long therapy-free interval is still under debate and needs to be balanced with medical treatment alone not only in relation to survival but with respect to a thorough evaluation of the quality of life. Practice points . proposal to include in the clinical management of advanced epithelial ovarian cancer patients a new integrated chemotherapeutic and surgical approach for those patients who cannot bene®t from primary cytoreductive surgery in achieving an `optimal' residual disease status . explain the background of the rationale for debulking surgery . include the data of those studies supporting the rationale to introduce interval debulking surgery in the context of the primary treatment of advanced epithelial ovarian cancer patients . include data of prospective clinical trials on interval debulking surgery
Research agenda . the value of interval debulking surgery in the management of advanced epithelial ovarian cancer patients who are not supposed to achieve an `optimal' residual disease status from primary cytoreductive surgery
REFERENCES 1. Grith CTH. Surgical resection of tumor bulk in the primary treatment of ovarian carcinoma. National Cancer Institute Monographs 1975; 42: 1010±1014. 2. Hoskins WJ et al. The eect of the diameter of largest residual disease on survival after primary cytoreductive surgery in patients with suboptimal residual epithelial ovarian carcinoma. American Journal of Obstetrics & Gynecology 1994; 170: 974±980. 3. Piver MS et al. The impact of aggressive debulking surgery and cisplatin based chemotherapy on progression-free survival in stage III and IV ovarian carcinoma. Journal of Clinical Oncology 1988; 6: 983±989.
582 S. Pecorelli, F. Odicino and G. Favalli 4. Heintz APM et al. The treatment of advanced ovarian carcinoma: clinical variables associated with prognosis. Gynecologic Oncology 1988; 30: 347±358. 5. Hacker NF et al. Primary cytoreductive surgery for epithelial ovarian cancer. Obstetrics & Gynecology 1983; 61: 413±420. 6. De Vita VT. Principles of chemotherapy. In De Vita VT et al (eds) Cancer: Principles and Practice of Oncology, 4th edn, pp 997±. Philadelphia, PA: Lippincott-Raven, 1997. 7. Grith CT et al. Role of cytoreductive surgical therapy in the management of advanced ovarian cancer. Cancer Treatment Report 1979; 63: 335±340. 8. Covens AL. A critique of surgical cytoreduction in advanced ovarian cancer. Gynecologic Oncology 2000; 78: 269±274. 9. Nejit JP, tenBokkel Huinink WW, van der Burg MEL et al. Randomized trial comparing two combination chemotherapy regimens (CHAP-5 v CP) in advanced ovarian carcinoma. Journal of Clinical Oncology 1987; 5: 1157±1168. 10. Jacob JH, Gersherson DM, Morris M et al. Neoadjuvant chemotherapy and interval debulking for advanced epithelial ovarian cancer. Gynecologic Oncology 1991; 42: 146±150. 11. Nejit JP, tenBokkel Huinink WW, van der Burg MEL et al. Long-term survival in ovarian cancer: mature data from the Netherland Joint Study Group for Ovarian Cancer. European Journal of Cancer 1991; 27: 1367±1372. 12. Nejit JP, tenBokkel Huinink WW, van den Burg MEL et al. Randomized trial comparing the combination chemotherapy regimens (CHAP-5 vs CP) in advanced ovarian carcinoma. Journal of Clinical Oncology 1987; 5: 1157±1168. 13. Varma R, Redman CWE, Blackledge G et al. Early second surgery in bulky epithelial ovarian cancer does not improve survival: results of a randomized study. Abstract of the 2nd meeting of IGCS, Toronto, October 9±13, 1989. 14. Redman CWE, Warwick J, Luesley DM et al. Intervention debulking surgery in advanced epithelial ovarian cancer. British Journal of Obstetrics and Gynaecology 1994; 101: 142±146. 15. van der Burg MEL, van Lent M, Buyse M et al. The eect of debulking surgery after induction chemotherapy on the prognosis in advanced epithelial ovarian cancer. New England Journal of Medicine 1995; 332: 629±634. 16. van der Burg MEL et al. After 6 years follow-up intervention debulking surgery remains an independent prognostic factor for survival in advanced epithelial ovarian cancer: an EORTC-GCCG study. International Journal of Gynecological Cancer 1997; 7: . 17. Kehoe S, Sha® M & Luesley D. Interval cytoreduction in ovarian cancer (letter). New England Journal of Medicine 1995; 333: 254. 18. Neijt JP. Interval cytoreduction in ovarian cancer (letter). New England Journal of Medicine 1995; 333: 254±255. 19. Cannistra SA. Interval cytoreduction in ovarian cancer (letter). New England Journal of Medicine 1995; 333: 255. 20. Van der Burg MEL, Buyse M & Pecorelli S. Interval cytoreduction in ovarian cancer (letter). New England Journal of Medicine 1995; 333: 255. 21. Blythe JG & Wahl TP. Debulking surgery: does it increase the quality of survival? Gynecologic Oncology 1982; 14: 396±408. 22. Hacker NF. Cytoreduction for advanced ovarian cancer in perspective. International Journal of Gynecological Cancer 1996; 6: 159±160. 23. Lawton FG et al. Neoadjuvant `cytoreductive' chemotherapy combined with intervention debulking surgery in advanced, unresected epithelial ovarian cancer. Obstetrics & Gynecolology 1989; 73: 61±65. 24. Surwit E et al. Neoadjuvant chemotherapy for advanced ovarian cancer. International Journal of Gynecological Cancer 1996; 6: 356±361. 25. Vergote I, De Wever I, Tjalma W et al. Neoadjuvant chemotherapy or primary debulking surgery in advanced ovarian carcinoma: a retrospective analysis of 285 patients. Gynecological Oncology 1998; 71: 431±436. 26. Ansquer Y, Leblanc E, Clough K et al. Neoadjuvant chemotherapy for unresectable ovarian carcinoma. Cancer 2001; 91: 2329±2334. 27. National Institute of Health Issue Consensus Development Conference Statement: Ovarian Cancer. Screening Treatment and Follow up. Gynecologic Oncology 1993; 4: 83±88. 28. Gore ME, Fryatt B, Wiltshaw & Dawson T. Treatment of relapsed carcinoma of the ovary with cisplatin or carboplatin following initial treatment with these compounds. Gynecologic Oncology 1990; 36: 207±211. 29. Markman M, Rothman R, Hakes T et al. Second line platinum therapy in patients with ovarian cancer previously treated with cisplatin. Journal of Clinical Oncology 1991; 9: 383±393. 30. Janicke F, Holscher M, Kuhn W et al. Radical surgical procedure improves survival time in patients with recurrent ovarian cancer. Cancer 1992; 70: 2129±2136.
Interval debulking surgery 583 31. Munkarah AR, Gershenson DM, Burke TE et al. Secondary cytoreductive surgery for a clinically apparent solitary intra-abdominal recurrence in patients with epithelial ovarian cancer. Proceedings of the American Society of Clinical Oncology 1996; 15: 295 (abstract). 32. Segna RA, Dottino PR, Mandeli JP et al. Secondary cytoreduction for ovarian cancer following cisplatin therapy. Journal of Clinical Oncology 1993; 11: 434±439. 33. Morris M, Gershenson DM, Taylor Wharton J et al. Secondary cytoreductive surgery for recurrent epithelial ovarian cancer. Gynecologic Oncology 1989; 34: 334±338. 34. Pecorelli S, Sartori E & Santin AD. Follow-up after primary therapy: management of the symptomatic patient. Surgery Gynecologic Oncology 1994; 55S: 138±142.