- Email: [email protected]
Critical evaluation of secondary cytoreduction in recurrent ovarian cancer
Gynecologic Oncology 95 (2004) 273 – 280 www.elsevier.com/locate/ygyno
Review
Critical evaluation of secondary cytoreduction in recurrent ovarian cancer A.R. Munkaraha,*, R.L. Colemanb a
Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Wayne State University, Karmanos Cancer Institute, Detroit, MI 48201, USA b Division of Gynecology Oncology, The University of Texas, M.D. Anderson Cancer Center, Houston, TX 77030, USA Received 15 September 2004
Abstract Objective. The optimal strategy for salvage therapy in patients who suffer from ovarian cancer recurrence after a disease-free interval is not established. The objective of this paper is to analyze the existing published data on salvage surgery in this setting. Methods. A retrospective review of the English literature was done looking at studies addressing the role of secondary cytoreductive surgery in recurrent ovarian cancer. A number of parameters were collected from these studies and analyzed, including patients’ characteristics, outcome of secondary cytoreduction, perioperative complications, postoperative therapy, and survival. In a parallel analysis, we reviewed the outcome of patients treated with salvage chemotherapy without surgery in similar clinical settings. Results. Optimal cytoreduction was achievable in 38–87% of the study populations reviewed with acceptable perioperative complications and mortality. The attempt to analyze the impact of secondary cytoreduction on survival was limited by (1) the inter-investigator differences in defining optimal cytoreduction, (2) the heterogeneity of the patients included, (3) and the lack of information on postoperative therapy. All the studies suggest that patients left with no gross residual disease after secondary cytoreduction seem to benefit from prolonged survival in the range of 44–60 months. Current data reveal that the use of combination chemotherapy without surgery for salvage treatment of recurrent ovarian cancer can be associated with prolonged median survival reaching up to 35 months. Conclusion. The available data suggest a benefit for secondary surgical cytoreduction in recurrent ovarian cancer. This needs to be considered in the light of recent data reporting prolonged survival with the use of combination salvage chemotherapy without surgery. Currently, it is not known if a salvage strategy combining surgery and multiagent chemotherapy regimens will have a survival benefit over chemotherapy alone. Hopefully, current ongoing prospective trials will answer this question. D 2004 Elsevier Inc. All rights reserved. Keywords: Secondary cytoreduction; Ovarian cancer; Chemotherapy
Contents Introduction . . . . . . . . . . . . . . . . . . . . . . Salvage secondary cytoreduction for recurrent disease Study populations . . . . . . . . . . . . . . . . . Outcome of secondary surgical cytoreduction . . . Survival data . . . . . . . . . . . . . . . . . . . . Prognostic factors . . . . . . . . . . . . . . . . . Resection of recurrences in the liver . . . . . . . . Limitations of the available literature . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
274 274 274 274 275 275 276 276
* Corresponding author. Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Wayne State University School of Medicine, Harper Professional Building, 4160 John R, Ste. 2135, 2nd Floor, Detroit, MI 48201. Fax: +1 313 993 8857. E-mail address: [email protected] (A.R. Munkarah). 0090-8258/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2004.09.018
274
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
Salvage second-line chemotherapy Single-agent platinum . . . . . Single-agent nonplatinum . . . Combination regimens . . . . . Conclusion . . . . . . . . . . . . References. . . . . . . . . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
Introduction Epithelial ovarian cancer continues to be one of the leading causes of cancer related deaths in the United States. According to the American Cancer Society estimates, 25,580 new cases will be diagnosed in 2004 and 16,090 women will die of the disease. One of the major problems with regard to ovarian cancer is the fact that 60% or more of the patients present with advanced stage disease. Despite significant improvement in the surgical care and the introduction of new and more efficacious chemotherapy combination, the majority of women diagnosed with advanced-stage cancer suffer from recurrence of their disease. Since therapy that can cure recurrent disease is essentially nonexistent, attempts at prolongation of overall and disease-free survival become an alternative goal. To achieve these endpoints, research is focusing on various alternatives that include new chemotherapy agents, surgical resection, as well as new biologic therapies targeting various cellular pathways. The question of secondary cytoreduction for recurrent ovarian cancer has been addressed in a number of publications that have been critiqued for a variety of reasons. The most important critiques focus on the fact that the studies are largely retrospective, span a long period of time, and are not randomized. As a result, there is significant heterogeneity in the treatments rendered as well as a high risk for selection bias. Despite that, one cannot ignore the consistent findings that, in women who underwent secondary cytoreduction, the size of residual disease significantly impacted survival. Opponents to the surgical approach for recurrent cancer argue that it is hard to determine whether the benefit reported is related to surgery or tumor biology. In this paper, we try to provide a detailed review of the available data regarding secondary cytoreduction. We also try to compare the patients’ outcome reported in those surgical trials to those reported in large phase II and III trials investigating second-line salvage chemotherapy without surgery.
Salvage secondary cytoreduction for recurrent disease Using a PubMed search of the literature, we limited our review to studies that looked at the impact of secondary cytoreduction exclusively in the setting of recurrent disease after a disease-free interval. The published studies were
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
276 277 277 277 278 279
thoroughly reviewed and the available data collected with respect to the surgical procedures performed, the rate of optimal surgical cytoreduction, the perioperative morbidity and mortality, the post-surgical cancer treatment, and overall survival. When mentioned, we analyzed the factors that influenced the success of surgical cytoreduction and patient’s survival. Study populations Twelve publications were identified that met our selection criteria [1–12]. Ten of these addressed the question of secondary cytoreduction in general [1–9,12]; the remaining two looked specifically at the issue of hepatic resections for recurrent ovarian cancer [10,11]. Since one might argue that recurrence of ovarian cancer in the liver is infrequent and might represent a biologically distinct disease, we decided to discuss the two related studies separately at the end of this section. The 10 other studies span a period of time between 1977 and 2001 (Table 1). The total umber of patients included in the nine studies was 631 with a range of 30–149 per study. The median age of the study populations ranged between 50 and 62 years. Seven of the studies were retrospective reviews; three were prospective nonrandomized studies where all patients presenting with recurrent disease were considered for a secondary cytoreductive surgery. In two studies, one by Eisenkop et al. [4] and the other by Scarabelli et al. [8], 114 and 177 patients with recurrent ovarian cancer were, respectively, considered for salvage surgery, of whom 106 and 149 actually underwent surgery. In the more recent study by Zang et al. [6], 117 patients underwent secondary tumor cytoreduction; however, it is not clear from the manuscript how many patients with recurrent cancer were prospectively evaluated for surgery. Outcome of secondary surgical cytoreduction In all the studies, various surgical procedures were performed including bowel resection, splenectomies, partial hepatic resections, and posterior exenteration. The mean operative times ranged between 133 and 300 min while the average blood loss ranged between 450 and 890 ml with 7000 ml being the maximum amount reported. Combining all the studies together, boptimalQ secondary cytoreduction was achievable in 67% (423/631) of patients [range: 38–87%; 95% confidence interval (CI): 63–71%]. The surgery was
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
275
Table 1 Outcome data on secondary cytoreductive surgery Author (year)
Morris (1989) Janick (1992) Vaccarello (1995) Eisenkop (2000) Gadduci (2000) Zang (2000) Munkarah (2001) Scarabelli (2001) Tay (2002) Zang (2004) a b
Number of patients
30 30 38 106 30 60 25 149 46 117
Definition of optimal debulking (cm)
%Patients optimally debulked
%Patients no gross residual tumor
Survival— optimally debulked group (months)
Survival— suboptimally debulked group (months)
Survival— group with no gross residual
b2 b2
60 87
13 47
18 9
– 29
b0.5 b0.5 – b1 b2 b1 b1 b1
39 85 83 38 72 70 72 62
13.3 All dead within 8 months 23 19.3a 19a 8 25.1 – 11 20
82 57 48 36 41 9
Not reached – – 19 56.9 – 14.5b 26b
– 44.4 37 – 100.1 – 38 Not reached
Median survival reported for women left with any macroscopic residual. Median survival for patients optimally debulked with gross residual V1 cm.
defined as optimal if the size of residual disease was b0.5 cm in two studies, b1 cm in four studies, and b2 cm in three studies; one study did not define the residual in the manuscript. Combining the data from eight studies that reported on complete surgical resection, all macroscopic disease was successfully resected in 41% (217/533) of patients (range: 13–82%; 95% CI: 37–45%). Major perioperative complications occurred in 11% (67/631) of patients (95% CI: 8–13%). These included cardiopulmonary events such as myocardial infarction and pulmonary emboli, major infections, wound dehiscence, and bowel injury. The perioperative mortality rate was 1.4% (range 0–3.4%). Survival data The overall survival after secondary cytoreduction for the study populations ranged between 16.3 and 35.9 months. It is hard to summarize the reported survival following secondary cytoreduction based on surgical outcome because of the significant heterogeneity in the analyses performed. Whereas some authors reported on survival of optimally vs. suboptimally debulked patients, others reported the survival based on the presence or absence of any macroscopic residual disease. Taking these limitations into account, we tried to be comprehensive in our analysis. The ranges for median survival of women based on residual disease after secondary cytoreduction were as follows: (1) 9–56.9 months for the optimal cytoreduction group; (2) and b8–25.1 months for the suboptimal group. In one study, the median survival for women left with b0.5 cm residual disease postsecondary cytoreduction was not reached at the time of publication; however, there was a 75% probability of surviving 41 months [3]. One group of patients that seemed to benefit the most from secondary cytoreduction included women who had complete resection of all macroscopic disease. Their median survival after salvage surgery in the various studies was significantly longer than any of the other groups and ranged between 29 and 100 months.
Looking at the three largest and prospective studies, the median survival for patients left with no macroscopic disease was 44 months in the study by Eisenkop et al. [4] and over 60 months for women who recurred later than 12 months after primary treatment and whose recurrent tumor was completely resected in the study by Scarabelli et al. [8]. The median survival for patients left with no gross residual disease was not reached yet at the time Zang et al. [6] submitted their publication; however, the authors reported that the 5-year survival rate for that specific group was 61.4%. Since postoperative therapy might impact survival, it is important to consider the various treatments patients received. Unfortunately, the detailed data on this issue is limited. In seven of the studies, some information regarding such therapy is available [1,2,4–6,8,10]. Out of the 487 patients in those studies, 333 received chemotherapy, 54 radiotherapy, and 7 hormonal treatment. Two studies reported that postoperative treatment had a significant influence on survival [2,6]. Prognostic factors In addition to analyzing the impact of residual disease on survival, various groups have looked at other factors that might impact prognosis. One factor that seemed to influence survival in most studies was the recurrence-free interval (RFI) prior to recurrence [2,5,8,9,12]. As suspected, a longer RFI was associated with a significantly longer survival after salvage surgery [5,8,9]. Some investigators reported that RFI impacted the outcome of salvage surgery and overall survival [2]. Others found that it did influence survival although it did not influence the surgical outcome with respect to the size of residual disease [8]. In a recent publication by Zang et al. [6], the time to relapse after completion of primary therapy had no impact on prognosis, a finding that contradicts an earlier study by the same group. In this more recent publication, the statistical analysis using
276
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
univariate analysis revealed that women with RFI N 24 months had the longest median survival of 40 months after secondary cytoreduction. However, the impact of RFI on survival was lost in the multivariate analysis. The authors argued that patients who recur as early as 3 months after primary therapy might benefit from a salvage surgery. Such conclusion is hardly supported in view of the small number of patients, the lack of data indicating how many of their patients were platinum-resistant and the fact that the median survival of women who recurred within 3–12 months was 18 months [1]. To put that in perspective, the median survival of all their patients with suboptimal cytoreduction was 20 months. Another variable that seems to influence resectability and survival is the number of sites of recurrence. In the study by Gadduci et al. [5], the number of recurrence sites had a significant influence on survival. Zang et al. [6] found that patients with localized recurrences had a significantly higher chance to achieve successful secondary cytoreduction and a longer survival. Eisenkop et al. [4] found that the number of sites did not influence success of surgery; however, the outcome of patients with diffuse peritoneal carcinomatosis was poor. Finally, some studies have noted that administration of salvage chemotherapy prior to salvage surgery might negatively influence survival [4,8]. However, these findings were not supported by other investigators [6,9].
38–70). The median recurrence-free interval was 36.5 months (range 6–243). The hepatic resection consisted of trisegmentectomy in 2 patients, lobectomy in 2, segmentectomy in 17, and wedge resection in 3. Resection of extrahepatic metastases was completed in 18 patients. All 24 patients were optimally debulked (b1 cm), 21 having complete resection of all macroscopic disease. Perioperative complications occurred in 5 patients, but there were no operative deaths. The median survival of their patients was 62 months (95% CI: 41–83). Limitations of the available literature There are a number of limitations in the current available literature on secondary cytoreduction. Most of the studies are retrospective and thus subject to significant selection bias. In the few prospective trials, this selection bias is diminished but not completely eliminated since no definite inclusion criteria are mentioned and there are no control groups. Optimal debulking is variously defined and thus comparing survival is very difficult. The recurrence-free interval prior to salvage surgery is also variable among the different studies. Finally, and most importantly, detailed data on salvage chemotherapy is not available. These data become extremely important as emerging evidence suggests that the type of salvage chemotherapy (platinum vs. nonplatinum drugs, single agent vs. combination) might have a significant impact on survival.
Resection of recurrences in the liver Salvage second-line chemotherapy Two recent groups published their experience with hepatic resection for recurrent ovarian cancer [10,11]. Meredith et al. reported on 26 patients who underwent hepatic resection for metachronous liver metastases from ovarian cancer. The median time between primary diagnosis of the ovarian cancer and the hepatic resection was 29.4 months (range 3.9–187.2). The median age of the study population was 62 years (range 39–75). The types of liver resection were segmentectomies in 18 cases, left hepatecttomy in 1, right hepatectomy in 4, and trisegmentectomies in 3. Twenty patients underwent simultaneous surgical resection of extrahepatic tumor. Three patients had significant perioperative complications but there were no perioperative deaths. Optimal cytoreduction was achieved in 21 patients, 18 of whom had no gross residual disease and 3 had residual nodules b1 cm in largest diameter. The median survival of the optimally debulked patients was 27.3 months compared to 8.6 months for the suboptimally debulked group. A time interval of less than 12 months between the primary and secondary cytoreduction had a negative impact on survival. On the other hand, neither the number nor the size of hepatic metastases had a prognostic influence. Yoon et al. [11] reported on 24 patients who underwent hepatic resection as part of their salvage surgery for recurrent ovarian (21 cases) or fallopian tube cancer (3 cases). The median age of their patient population was 53 years (range
Although most women with advanced disease at presentation ultimately recur, they generally harbor good performance status and are emotionally accepting of secondary therapeutic considerations. Whether deemed appropriate for surgery or not, most will be considered for systemic chemotherapy and many will receive more than two regimens in the salvage setting. To understand whether surgery, as outlined above, provides bvalue-addedQ survival when considered with chemotherapy, it is important to review trials of patients similarly selected but treated by chemotherapy alone. It is clear that patient selection bias prohibits direct comparison to patients evaluated in the aforementioned surgical trials. It is equally problematic to report results from individual phase II trials without regard to the important prognostic factors that appear to govern response, such as treatment-free interval (TFI) and prior platinum and taxane exposure. Therefore, in this section, we present data from relevant phase II and phase III trials with respect to these important prognosticators (particularly platinum sensitivity and treatment-free survival) in order to provide some relative parity for context. The options of salvage chemotherapy for recurrent ovarian cancer include treatment with (1) single-agent platinum compound, (2) single-agent nonplatinum drug,
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
(3) platinum-based combination regimen, and (4) nonplatinum-based combination regimen. Single-agent platinum Since the platinum compounds are generally regarded as the most active agents in advanced ovarian cancer, platinum-based therapy is often considered first when platinum-sensitive disease recurs. Patients who have a prolonged duration of response to first-line platinum-based therapy have an increased likelihood of response to platinum retreatment [13,14]. Combined data from the landmark Gore and Markman studies show increased response rates with increasing TFI up to 24 months and beyond; in fact, the response rate to a platinum agent after a TFI of 5–12 months is 22%; compared to 31% and 59%, respectively, for TFI of 12–24 and N24 months. These trials were largely conducted in an era of relatively few effective chemotherapeutic alternatives for women with recurrent ovarian cancer. Nonetheless, a study confirming these observations was recently presented by Markman et al. who retreated 214 platinum-sensitive patients with predominately single-agent carboplatin. Responses were seen overall in 55% of patients but this was significantly influenced by TFI. Those patients with a TFI of less than 12 months had a 35% response rate; those with TFI greater than 36 months had a response rate of 80%. Single-agent nonplatinum Similar bTFI-relatedQ effects are also seen with many of the novel nonplatinum agents. For instance, in the large randomized trial of recurrent ovarian cancer patients treated with either single-agent liposomal doxorubicin or topotecan, clinical response rates were 12% and 7%, respectively, in 254 platinum-resistant patients (defined as progression on or recurrence within 6 months of front-line therapy), and 28% and 29%, respectively among 220 platinum-sensitive patients [15]. Survival was similarly influenced in this trial, being a median 36 and 41 weeks, respectively, in platinumresistant patients treated with liposomal doxorubicin or topotecan, and 108 and 71 weeks in platinum-sensitive patients. Though debatable, some investigators have reported that progression-free survival may be the most important factor in determining subsequent tumor response and survival. In a review of eight GINECO Phase II and III chemotherapy clinical trials involving 583 patients in the recurrent setting, Pujade-Lauraine et al. [16] demonstrated that response and survival were closely related to initial relapse-free interval. Patients recurring in 3–12 months of their initial treatment had a response rate of 35% and an overall survival of 393 days; in the cohort with a relapse-free survival of at least 18 months, response was 62% and overall survival was 957 days. Importantly, these investigators found these effects to be independent of the individual chemotherapeutic agents.
277
This positive influence of a longer TFI on patients’ survival is similar to what has been described in the secondary cytoreduction studies. Thus, it would be reasonable to expect that patients generally considered for secondary surgery followed by chemotherapy would enjoy by selection, a longer survivorship than those with shorter TFI treated with chemotherapy alone. Combination regimens Another popular treatment strategy in patients with long TFI is the use of combination chemotherapy. Some clinicians maintain that a combination regimen may be better than single-agent carboplatin or cisplatin, since some phase II data suggest somewhat higher response rates and disease-free survivals with combination therapy. Markman et al. [17] made the observation that in their patients’ secondary TFI were temporally shorter than the initial TFI; however, the reverse was seen in three patients who received combination salvage chemotherapy suggesting a beneficial role for combination chemotherapy in the salvage setting. Dizon et al. [18] reported a 70% overall response rate (42% CR) among 66 drug-sensitive patients (median treatmentfree interval 22 months) with combination paclitaxel and carboplatin. The median progression-free interval following therapy was 13 months and 3-year overall survival was 72%. In order to place these data into perspective against a single-agent approach, these same authors recently reported data on a similar cohort treated with single-agent carboplatin [19]. All response parameters were significantly lower among the 22 patients undergoing single-agent therapy including response rates (36% vs. 70%) and median survival (26 vs. 42 months). The median treatment-free interval was 22 months. These favorable responses must be carefully interpreted against their study cohort, which was significantly heterogeneous. It is clear that combination therapy in platinum-sensitive patients is being increasingly evaluated in phase II clinical trials. Most are incorporating agents that have demonstrated historical activity in ovarian cancer or have the potential to demonstrate additive or synergistic activity. One trial highlighting this latter point was reported by Nagourney et al. [20] studying the gemcitabine/cisplatin combination in a mixed group of recurrent ovarian cancer patients (n = 27). Evidence for synergistic activity between these agents is well documented and appears to be related not only to cisplatin’s ability to inhibit ribonucleotide reductase, thereby reducing the rate-limiting inhibition of gemcitabine phosphorylation, but also through enhanced inhibition of the repair of cisplatin-induced DNA adducts. In this small trial, 11 of 13 (84%) patients with a median platinum-free interval of 12 months responded including four complete responders. Overall survival for the study population was 20.9 months. Prospective data comparing single-agent to combination chemotherapy are very limited (Table 2). Bolis et al. [21] reported on 190 platinum-sensitive (N6 months) patients who
278
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
Table 2 Comparative studies on salvage chemotherapy Agent/combination chemotherapy [references]
N
% platinumsensitivea
TFI (months)
PFS (weeks)
P
OS (weeks)
P
Topotecan vs. paclitaxel [24] Paclitaxel vs. weekly paclitaxel [25] Oxaliplatin vs. paclitaxel [26] DoxilR vs. topotecan [15] Topotecan vs. treosulfan [27] Paclitaxel vs. paclitaxel/epirubicin [28] Carboplatin vs. carboplatin/epirubicin [21] Paclitaxel vs. CAP [29] Platinum regimen vs. platinum/paclitaxel [22] Carboplatin vs. carboplatin/gemcitabine [23]
226 208 86 481 357 81 190 97 802 356
47 48 26 46 Mixed – – – – –
– – – – – 1.5 17 34.5 N12 N6
23 38 12 16 22 26 65 39 43 25
NS NS NS NS 0.001 NS NS 0.04 b0.001 0.004
61 vs. 64 vs. 42 vs. 60 vs. 56 vs. 39 vs. 109 vs. 112 vs. 104 vs. –
NS NS NS NS 0.02 NS NS 0.04 0.023 –
vs. vs. vs. vs. vs. vs. vs. vs. vs. vs.
14 26 14 17 12 43 78 68 52 37
43 59 37 57 48 61 122 151 130
TFI: treatment-free interval; PFS: progression-free survival; OS: overall survival. a Percent of the platinum-sensitive patients in the study population.
were randomized to receive either single agent carboplatin (300 mg/m2) or combination carboplatin (300 mg/m2) and epirubicin (120 mg/m2). The median platinum-free interval was 17 months and a minimum of five cycles of therapy was planned. The overall response rate was 55% for carboplatin vs. 58% for the combination. Median progression-free and overall survivals favored the combination regimen but were not statistically significant being 15 and 25 months in the carboplatin arm and 18 and 28 months in the carboplatin/ epirubicin arm, respectively. Limitation in sample size may have contributed to the effect. Toxicity, however, was significantly increased, with hematological and gastrointestinal toxicities occurring more frequently in the combination cohort. Recently, the ICON collaborators conducted a large randomized trial of paclitaxel and platinum vs. platinumbased conventional therapy in first relapse platinum-sensitive patients [22]. This 802 patient trial represents the largest such trial completed to date in this setting. Patients were allowed entry following primary platinum-based chemotherapy and were progression-free a minimum of 6 months in patients recruited from the Medical Council’s Clinical Trials Unit (MRC) and Arbeitsgemeinschaft Gynaekologische Onkologie (AGO) and a minimum 12 months in those recruited from the Instituto Mario Negri (IRFMN). Overall response was 66% for the combination and 54% for the single agent ( P = 0.06). A significant improvement in progression-free (9 vs. 12 months, HR = 0.76, P b 0.001) and overall survival (24 vs. 29 months, HR = 0.82, P = 0.023) were observed for the combination of paclitaxel and platinum at a median follow-up of 42 months. Toxicity was mixed between the arms with more hematologic toxicity seen in the conventional platinum arm particularly singleagent carboplatin, and more neurotoxicity was observed in the taxane combination regiments. This provocative data, if confirmed, will likely justify combination paclitaxel and platinum chemotherapy in the cohort of patients also strongly considered for secondary surgery, and may be the most promising strategy following a salvage surgical intervention. A second prospective randomized trial conducted by the Gynecologic Cancer Intergroup (GCIG) has
been recently reported [23]. In this trial, 356 patients with recurrent ovarian cancer were allocated to either single agent carboplatin or combination carboplatin and gemcitabine. Currently, data on the time to progression endpoint confirm an improved activity for the combination regimen. In the trial, the hazard for progression after treatment was 0.72 (95% CI: 0.6–0.9) amounting to a 12-week increase in progression-free interval (23 vs. 35 wks). Response rates were also higher in the combination (47% vs. 31%, P = 0.0016). As expected toxicity, predominately hematologic, was more common in the combination arm. The unique combination, which may highlight intrinsic drug synergy, illustrates that other combinations may also be of interest in this setting.
Conclusion Although marginally less biased than the aforementioned surgical trials, the collective experience from the chemotherapy studies would suggest that equally selected platinum-sensitive patients undergoing salvage chemotherapy with no surgery may be expected to have durable survival. In the overall experience from the presented phase II, randomized single-agent and randomized combination chemotherapy trials, overall survival ranged between 24 and 42 months. However, none of the chemotherapy trials reports median survivals that are as prolonged as those seen in patients who undergo complete resection of tumor recurrence and receive adjuvant salvage chemotherapy. One can argue that a longer progression-free interval after primary therapy reflects a tumor that is biologically less aggressive and more chemoresponsive. Therefore, patients who have recurrent disease after a prolonged progressionfree interval might benefit most from a combined approach at complete tumor cytoreduction and chemotherapy similar to the approach adopted in primary treatment. It is clear that the available literature does not offer the answer as to how patients should be selected for secondary cytoreduction, who would benefit from such surgical
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
279
Fig. 1. Proposed schema for protocol 213.
approach, and what impact this will have on their quality of life; this will only be determined from carefully controlled randomized trials. Given the nature of such trials and the number of prognostic factors such as patient age and performance status, disease stage, length of progressionfree interval, and extent of recurrent disease, these studies will be necessarily large and ambitious. In addition, the recent data suggesting the added survival benefit of salvage combination chemotherapy necessitate a careful analysis of the types of chemotherapy regimens delivered after surgery. Since the benefit of the latter has only been demonstrated in one randomized trial, balancing the effect of two aggressive approaches, surgery and combination chemotherapy, with quality of life will be important in order to provide the context with which to interpret any observed benefit. Currently, one such trial from the EORTC (Protocol# 55963) is underway and accruing patients. The inclusion criteria from this trial are disease recurrence, after a minimum of three cycles of primary chemotherapy and—a disease-free interval z12 months. Patients are randomized to either six cycles of single-agent platinum (carboplatin or cisplatin) or three cycles of the same chemotherapy, followed by surgical exploration and three more cycles of chemotherapy. They are stratified by initial stage, progression-free survival (b24 vs. N24 months), initial response to primary therapy, and tumor burden. Anticipated accrual is 700 patients and survival is the primary endpoint. A second trial, shown in Fig. 1, is being conducted through the GOG (Protocol #213). This trial randomizes recurrent ovarian and peritoneal patients with a minimum treatment-free interval of 6 months to either surgical exploration or chemotherapy. Given the provocative data presented above, a secondary randomization is being
made between platinum-based combination chemotherapy, combination of chemotheraphy and biologics and a control arm defined as single agent carboplatin followed by a paclitaxel in third line. It is hoped that those trials, when completed, will provide more conclusive data as to the optimal management of patients who suffer from recurrent ovarian cancer after a complete remission.
References [1] Morris M, Gershenson M, Wharton T, Copeland LJ, Edwards CL, Stringer CA. Secondary cytoreductive surgery for recurrent epithelial ovarian cancer. Gynecol Oncol 1989;34:334 – 8. [2] Janicke F, Holscher M, Kuhn W, Hugo RV, Pache L, Siewert JR, et al. Radical surgical procedure improves survival time in patients with recurrent ovarian cancer. Cancer 1992;70. [3] Vaccarello L, Rubin SC, Vlamis VGW, Jones WB, Lewis JL, Hoskins WJ. Cytoreductive surgery in ovarian carcinoma patients with a documented previously complete surgical response. Gynecol Oncol 1995;57:61 – 5. [4] Eisenkop SM, Friedman RL, Spirtos NM. The role of secondary cytoreductive surgery in the treatment of patients with recurrent epithelial ovarian carcinoma. Cancer 2000;88. [5] Gadducci A, Iacconi P, Cosio S, Fanucchi A, Cristofani R, Genazzani AR. Complete salvage surgical cytoreduction improves further survival of patients with late recurrent ovarian cancer. Gynecol Oncol 2000;79:344 – 9. [6] Zang RY, Li ZT, Tang J, Cheng X, Cai SM, Zhang ZY, et al. Secondary cytoreductive surgery for patients with relapsed epithelial ovarian carcinoma: who benefits? Cancer 2004;100:1152 – 61. [7] Munkarah AR, Levenback C, Wolf JK, Bodurka-Bevers D, TortoleroLuna G, Morris R, et al. Secondary cytoreductive surgery for localized intra-abdominal recurrences in epithelial ovarian cancer. Gynecol Oncol 2001;81:237 – 41.
280
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
[8] Scarabelli C, Gallo A, Carbone A. Secondary cytoreductive surgery for patients with recurrent epithelial ovarian carcinoma. Gynecol Oncol 2001;83:504 – 12. [9] Tay EH, Grant PT, Gebski V, Hacker NF. Secondary cytoreductive surgery for recurrent epithelial ovarian cancer. Obstet Gynecol 2002;99. [10] Merideth MA, Cliby WA, Keeney GL, Lesnick TG, Nagorney DM, Podratz KC. Hepatic resection for metachronous metastases from ovarian carcinoma. Gynecol Oncol 2003. [11] Yoon SS, Jarnagin WR, Fong Y, DeMatteo RP, Barakat RR, Blumgart LH, et al. Resection of recurrent ovarian or fallopian tube carcinoma involving the liver. Gynecol Oncol 2003;91:383 – 8. [12] Zang RY, Zhang ZY, Li ZT, Chen J, Tang MQ, Liu Q, et al. Effect of cytoreductive surgery on survival of patients with recurrent epithelial ovarian cancer. J Surg Oncol 2000;75:24 – 30. [13] Markman M, Rothman R, Hakes T. Second-line platinum therapy in patients with ovarian cancer previously treated with cisplatin. J Clin Oncol 1991;9:389 – 93. [14] Gore ME, Fryatt I, Wiltshaw E. Cisplatin/carboplatin cross-resistance in ovarian cancer. Br J Cancer 1989;60:767 – 9. [15] Gordon AN, Fleagle JT, Guthrie D. Recurrent epithelial ovarian carcinoma: a randomized phase II study of pegylated liposomal doxorubicin versus topotecan. J Clin Oncol 2001;19:3312 – 22. [16] Pujade-Lauraine E, Paraiso D, Cure H. Predicting the effectiveness of chemotherapy (CX) in patients with recurrent ovarian cancer (ROC): a GINECO study: Abst#829. Proc Am Soc Clin Oncol 2002. [17] Markman M, Markman JR, Zanotti KM. Duration of response to second-line platinum-based chemotherapy for ovarian cancer: implication for patient management and clinical trial design. Proc Am Soc Clin Oncol 2003;22:447. [18] Dizon DS, Hensley ML, Poynor EA. Retrospective analysis of carboplatin and paclitaxel as initial second-line therapy for recurrent epithelial ovarian carcinoma: application toward a dynamic disease state of model of ovarian cancer. J Clin Oncol 2002;20:1238 – 47. [19] Dizon DS, Dupont J, Anderson S. Treatment of recurrent ovarian cancer: a retrospective analysis of women treated with single-agent carboplatin originally treated with carboplatin and paclitaxel. The Memorial Sloan-Kettering Cancer Center experience. Gynecol Oncol 2003;91:584 – 90.
[20] Nagourney RA, Brewer CA, Radecki S. Phase II trial of gemcitabine plus cisplatin repeating doublet therapy in previously treated, relapsed ovarian cancer patients. Gynecol Oncol 2003;88:35 – 9. [21] Bolis G, Scarfone G, Giardina G. Carboplatin alone vs. carboplatin plus epidoxorubicin as second-line therapy for cisplatin- or carboplatin-sensitive ovarian cancer. Gynecol Oncol 2001;81:3 – 9. [22] Parmar MK, Ledermann JA, Colombo N. Paclitaxel plus platinumbased chemotherapy versus conventional platinum-based chemotherapy in women with relapsed ovarian cancer: the ICON4/AGOOVAR-2.2 trial. Lancet 2003;361:2099 – 106. [23] Pfisterer J, Plante M, Vergote I. Gemcitabine/carboplatin vs. carboplatin in platinum sensitive recurrent ovarian cancer. Results of a Gynecologic Cancer Intergroup randomized phase III trial of the AGO OVAR, the NCIC CTG, and the EORTC GCG. Proc Am Soc Clin Oncol 2004;23:449. [24] ten Bokkel Huinink W, Gore M, Carmichael J. Topotecan versus paclitaxel for the treatment of recurrent epithelial ovarian cancer. J Clin Oncol 1997;15:2183 – 93. [25] Rosenberg P, Anderson H, Boman K. Randomized trial of single agent paclitaxel given weekly versus every three weeks and with peroral versus intravenous steroid premedication to patients with ovarian cancer previously treated with platinum. Acta Oncol 2002;41:418 – 24. [26] Piccart MJ, Green JA, Lacave AJ. Oxaliplatin or paclitaxel in patients with platinum-pretreated advanced ovarian cancer: a randomized phase II study of the European Organization for Research and Treatment of Cancer Gynecology Group. J Clin Oncol 2000;18: 1193 – 202. [27] Meier W, du Bois A, Kuhn W. Topotecan vs. treosulfan in recurrent ovarian cancer after initial therapy with platinum and paclitaxel. A prospective randomized phase III study. Proc Am Soc Clin Oncol 2003;22:450. [28] Bolis G, Parazzini F, Scarfone G. Paclitaxel vs. epidoxorubicin plus paclitaxel as second-line therapy for platinum-refractory and-resistant ovarian cancer. Gynecol Oncol 1999;72:60 – 4. [29] Cantu MG, Buda A, Parma G, Rossi R, Floriani I, Bonazzi C, et al. Randomized controlled trial of single-agent paclitaxel versus cyclophosphamide, doxorubicin, and cisplatin in patients with recurrent ovarian cancer who responded to first-line platinum-based regimens. J Clin Oncol 2002;5:1158 – 60.
Review
Critical evaluation of secondary cytoreduction in recurrent ovarian cancer A.R. Munkaraha,*, R.L. Colemanb a
Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Wayne State University, Karmanos Cancer Institute, Detroit, MI 48201, USA b Division of Gynecology Oncology, The University of Texas, M.D. Anderson Cancer Center, Houston, TX 77030, USA Received 15 September 2004
Abstract Objective. The optimal strategy for salvage therapy in patients who suffer from ovarian cancer recurrence after a disease-free interval is not established. The objective of this paper is to analyze the existing published data on salvage surgery in this setting. Methods. A retrospective review of the English literature was done looking at studies addressing the role of secondary cytoreductive surgery in recurrent ovarian cancer. A number of parameters were collected from these studies and analyzed, including patients’ characteristics, outcome of secondary cytoreduction, perioperative complications, postoperative therapy, and survival. In a parallel analysis, we reviewed the outcome of patients treated with salvage chemotherapy without surgery in similar clinical settings. Results. Optimal cytoreduction was achievable in 38–87% of the study populations reviewed with acceptable perioperative complications and mortality. The attempt to analyze the impact of secondary cytoreduction on survival was limited by (1) the inter-investigator differences in defining optimal cytoreduction, (2) the heterogeneity of the patients included, (3) and the lack of information on postoperative therapy. All the studies suggest that patients left with no gross residual disease after secondary cytoreduction seem to benefit from prolonged survival in the range of 44–60 months. Current data reveal that the use of combination chemotherapy without surgery for salvage treatment of recurrent ovarian cancer can be associated with prolonged median survival reaching up to 35 months. Conclusion. The available data suggest a benefit for secondary surgical cytoreduction in recurrent ovarian cancer. This needs to be considered in the light of recent data reporting prolonged survival with the use of combination salvage chemotherapy without surgery. Currently, it is not known if a salvage strategy combining surgery and multiagent chemotherapy regimens will have a survival benefit over chemotherapy alone. Hopefully, current ongoing prospective trials will answer this question. D 2004 Elsevier Inc. All rights reserved. Keywords: Secondary cytoreduction; Ovarian cancer; Chemotherapy
Contents Introduction . . . . . . . . . . . . . . . . . . . . . . Salvage secondary cytoreduction for recurrent disease Study populations . . . . . . . . . . . . . . . . . Outcome of secondary surgical cytoreduction . . . Survival data . . . . . . . . . . . . . . . . . . . . Prognostic factors . . . . . . . . . . . . . . . . . Resection of recurrences in the liver . . . . . . . . Limitations of the available literature . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
274 274 274 274 275 275 276 276
* Corresponding author. Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Wayne State University School of Medicine, Harper Professional Building, 4160 John R, Ste. 2135, 2nd Floor, Detroit, MI 48201. Fax: +1 313 993 8857. E-mail address: [email protected] (A.R. Munkarah). 0090-8258/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2004.09.018
274
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
Salvage second-line chemotherapy Single-agent platinum . . . . . Single-agent nonplatinum . . . Combination regimens . . . . . Conclusion . . . . . . . . . . . . References. . . . . . . . . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
Introduction Epithelial ovarian cancer continues to be one of the leading causes of cancer related deaths in the United States. According to the American Cancer Society estimates, 25,580 new cases will be diagnosed in 2004 and 16,090 women will die of the disease. One of the major problems with regard to ovarian cancer is the fact that 60% or more of the patients present with advanced stage disease. Despite significant improvement in the surgical care and the introduction of new and more efficacious chemotherapy combination, the majority of women diagnosed with advanced-stage cancer suffer from recurrence of their disease. Since therapy that can cure recurrent disease is essentially nonexistent, attempts at prolongation of overall and disease-free survival become an alternative goal. To achieve these endpoints, research is focusing on various alternatives that include new chemotherapy agents, surgical resection, as well as new biologic therapies targeting various cellular pathways. The question of secondary cytoreduction for recurrent ovarian cancer has been addressed in a number of publications that have been critiqued for a variety of reasons. The most important critiques focus on the fact that the studies are largely retrospective, span a long period of time, and are not randomized. As a result, there is significant heterogeneity in the treatments rendered as well as a high risk for selection bias. Despite that, one cannot ignore the consistent findings that, in women who underwent secondary cytoreduction, the size of residual disease significantly impacted survival. Opponents to the surgical approach for recurrent cancer argue that it is hard to determine whether the benefit reported is related to surgery or tumor biology. In this paper, we try to provide a detailed review of the available data regarding secondary cytoreduction. We also try to compare the patients’ outcome reported in those surgical trials to those reported in large phase II and III trials investigating second-line salvage chemotherapy without surgery.
Salvage secondary cytoreduction for recurrent disease Using a PubMed search of the literature, we limited our review to studies that looked at the impact of secondary cytoreduction exclusively in the setting of recurrent disease after a disease-free interval. The published studies were
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
276 277 277 277 278 279
thoroughly reviewed and the available data collected with respect to the surgical procedures performed, the rate of optimal surgical cytoreduction, the perioperative morbidity and mortality, the post-surgical cancer treatment, and overall survival. When mentioned, we analyzed the factors that influenced the success of surgical cytoreduction and patient’s survival. Study populations Twelve publications were identified that met our selection criteria [1–12]. Ten of these addressed the question of secondary cytoreduction in general [1–9,12]; the remaining two looked specifically at the issue of hepatic resections for recurrent ovarian cancer [10,11]. Since one might argue that recurrence of ovarian cancer in the liver is infrequent and might represent a biologically distinct disease, we decided to discuss the two related studies separately at the end of this section. The 10 other studies span a period of time between 1977 and 2001 (Table 1). The total umber of patients included in the nine studies was 631 with a range of 30–149 per study. The median age of the study populations ranged between 50 and 62 years. Seven of the studies were retrospective reviews; three were prospective nonrandomized studies where all patients presenting with recurrent disease were considered for a secondary cytoreductive surgery. In two studies, one by Eisenkop et al. [4] and the other by Scarabelli et al. [8], 114 and 177 patients with recurrent ovarian cancer were, respectively, considered for salvage surgery, of whom 106 and 149 actually underwent surgery. In the more recent study by Zang et al. [6], 117 patients underwent secondary tumor cytoreduction; however, it is not clear from the manuscript how many patients with recurrent cancer were prospectively evaluated for surgery. Outcome of secondary surgical cytoreduction In all the studies, various surgical procedures were performed including bowel resection, splenectomies, partial hepatic resections, and posterior exenteration. The mean operative times ranged between 133 and 300 min while the average blood loss ranged between 450 and 890 ml with 7000 ml being the maximum amount reported. Combining all the studies together, boptimalQ secondary cytoreduction was achievable in 67% (423/631) of patients [range: 38–87%; 95% confidence interval (CI): 63–71%]. The surgery was
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
275
Table 1 Outcome data on secondary cytoreductive surgery Author (year)
Morris (1989) Janick (1992) Vaccarello (1995) Eisenkop (2000) Gadduci (2000) Zang (2000) Munkarah (2001) Scarabelli (2001) Tay (2002) Zang (2004) a b
Number of patients
30 30 38 106 30 60 25 149 46 117
Definition of optimal debulking (cm)
%Patients optimally debulked
%Patients no gross residual tumor
Survival— optimally debulked group (months)
Survival— suboptimally debulked group (months)
Survival— group with no gross residual
b2 b2
60 87
13 47
18 9
– 29
b0.5 b0.5 – b1 b2 b1 b1 b1
39 85 83 38 72 70 72 62
13.3 All dead within 8 months 23 19.3a 19a 8 25.1 – 11 20
82 57 48 36 41 9
Not reached – – 19 56.9 – 14.5b 26b
– 44.4 37 – 100.1 – 38 Not reached
Median survival reported for women left with any macroscopic residual. Median survival for patients optimally debulked with gross residual V1 cm.
defined as optimal if the size of residual disease was b0.5 cm in two studies, b1 cm in four studies, and b2 cm in three studies; one study did not define the residual in the manuscript. Combining the data from eight studies that reported on complete surgical resection, all macroscopic disease was successfully resected in 41% (217/533) of patients (range: 13–82%; 95% CI: 37–45%). Major perioperative complications occurred in 11% (67/631) of patients (95% CI: 8–13%). These included cardiopulmonary events such as myocardial infarction and pulmonary emboli, major infections, wound dehiscence, and bowel injury. The perioperative mortality rate was 1.4% (range 0–3.4%). Survival data The overall survival after secondary cytoreduction for the study populations ranged between 16.3 and 35.9 months. It is hard to summarize the reported survival following secondary cytoreduction based on surgical outcome because of the significant heterogeneity in the analyses performed. Whereas some authors reported on survival of optimally vs. suboptimally debulked patients, others reported the survival based on the presence or absence of any macroscopic residual disease. Taking these limitations into account, we tried to be comprehensive in our analysis. The ranges for median survival of women based on residual disease after secondary cytoreduction were as follows: (1) 9–56.9 months for the optimal cytoreduction group; (2) and b8–25.1 months for the suboptimal group. In one study, the median survival for women left with b0.5 cm residual disease postsecondary cytoreduction was not reached at the time of publication; however, there was a 75% probability of surviving 41 months [3]. One group of patients that seemed to benefit the most from secondary cytoreduction included women who had complete resection of all macroscopic disease. Their median survival after salvage surgery in the various studies was significantly longer than any of the other groups and ranged between 29 and 100 months.
Looking at the three largest and prospective studies, the median survival for patients left with no macroscopic disease was 44 months in the study by Eisenkop et al. [4] and over 60 months for women who recurred later than 12 months after primary treatment and whose recurrent tumor was completely resected in the study by Scarabelli et al. [8]. The median survival for patients left with no gross residual disease was not reached yet at the time Zang et al. [6] submitted their publication; however, the authors reported that the 5-year survival rate for that specific group was 61.4%. Since postoperative therapy might impact survival, it is important to consider the various treatments patients received. Unfortunately, the detailed data on this issue is limited. In seven of the studies, some information regarding such therapy is available [1,2,4–6,8,10]. Out of the 487 patients in those studies, 333 received chemotherapy, 54 radiotherapy, and 7 hormonal treatment. Two studies reported that postoperative treatment had a significant influence on survival [2,6]. Prognostic factors In addition to analyzing the impact of residual disease on survival, various groups have looked at other factors that might impact prognosis. One factor that seemed to influence survival in most studies was the recurrence-free interval (RFI) prior to recurrence [2,5,8,9,12]. As suspected, a longer RFI was associated with a significantly longer survival after salvage surgery [5,8,9]. Some investigators reported that RFI impacted the outcome of salvage surgery and overall survival [2]. Others found that it did influence survival although it did not influence the surgical outcome with respect to the size of residual disease [8]. In a recent publication by Zang et al. [6], the time to relapse after completion of primary therapy had no impact on prognosis, a finding that contradicts an earlier study by the same group. In this more recent publication, the statistical analysis using
276
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
univariate analysis revealed that women with RFI N 24 months had the longest median survival of 40 months after secondary cytoreduction. However, the impact of RFI on survival was lost in the multivariate analysis. The authors argued that patients who recur as early as 3 months after primary therapy might benefit from a salvage surgery. Such conclusion is hardly supported in view of the small number of patients, the lack of data indicating how many of their patients were platinum-resistant and the fact that the median survival of women who recurred within 3–12 months was 18 months [1]. To put that in perspective, the median survival of all their patients with suboptimal cytoreduction was 20 months. Another variable that seems to influence resectability and survival is the number of sites of recurrence. In the study by Gadduci et al. [5], the number of recurrence sites had a significant influence on survival. Zang et al. [6] found that patients with localized recurrences had a significantly higher chance to achieve successful secondary cytoreduction and a longer survival. Eisenkop et al. [4] found that the number of sites did not influence success of surgery; however, the outcome of patients with diffuse peritoneal carcinomatosis was poor. Finally, some studies have noted that administration of salvage chemotherapy prior to salvage surgery might negatively influence survival [4,8]. However, these findings were not supported by other investigators [6,9].
38–70). The median recurrence-free interval was 36.5 months (range 6–243). The hepatic resection consisted of trisegmentectomy in 2 patients, lobectomy in 2, segmentectomy in 17, and wedge resection in 3. Resection of extrahepatic metastases was completed in 18 patients. All 24 patients were optimally debulked (b1 cm), 21 having complete resection of all macroscopic disease. Perioperative complications occurred in 5 patients, but there were no operative deaths. The median survival of their patients was 62 months (95% CI: 41–83). Limitations of the available literature There are a number of limitations in the current available literature on secondary cytoreduction. Most of the studies are retrospective and thus subject to significant selection bias. In the few prospective trials, this selection bias is diminished but not completely eliminated since no definite inclusion criteria are mentioned and there are no control groups. Optimal debulking is variously defined and thus comparing survival is very difficult. The recurrence-free interval prior to salvage surgery is also variable among the different studies. Finally, and most importantly, detailed data on salvage chemotherapy is not available. These data become extremely important as emerging evidence suggests that the type of salvage chemotherapy (platinum vs. nonplatinum drugs, single agent vs. combination) might have a significant impact on survival.
Resection of recurrences in the liver Salvage second-line chemotherapy Two recent groups published their experience with hepatic resection for recurrent ovarian cancer [10,11]. Meredith et al. reported on 26 patients who underwent hepatic resection for metachronous liver metastases from ovarian cancer. The median time between primary diagnosis of the ovarian cancer and the hepatic resection was 29.4 months (range 3.9–187.2). The median age of the study population was 62 years (range 39–75). The types of liver resection were segmentectomies in 18 cases, left hepatecttomy in 1, right hepatectomy in 4, and trisegmentectomies in 3. Twenty patients underwent simultaneous surgical resection of extrahepatic tumor. Three patients had significant perioperative complications but there were no perioperative deaths. Optimal cytoreduction was achieved in 21 patients, 18 of whom had no gross residual disease and 3 had residual nodules b1 cm in largest diameter. The median survival of the optimally debulked patients was 27.3 months compared to 8.6 months for the suboptimally debulked group. A time interval of less than 12 months between the primary and secondary cytoreduction had a negative impact on survival. On the other hand, neither the number nor the size of hepatic metastases had a prognostic influence. Yoon et al. [11] reported on 24 patients who underwent hepatic resection as part of their salvage surgery for recurrent ovarian (21 cases) or fallopian tube cancer (3 cases). The median age of their patient population was 53 years (range
Although most women with advanced disease at presentation ultimately recur, they generally harbor good performance status and are emotionally accepting of secondary therapeutic considerations. Whether deemed appropriate for surgery or not, most will be considered for systemic chemotherapy and many will receive more than two regimens in the salvage setting. To understand whether surgery, as outlined above, provides bvalue-addedQ survival when considered with chemotherapy, it is important to review trials of patients similarly selected but treated by chemotherapy alone. It is clear that patient selection bias prohibits direct comparison to patients evaluated in the aforementioned surgical trials. It is equally problematic to report results from individual phase II trials without regard to the important prognostic factors that appear to govern response, such as treatment-free interval (TFI) and prior platinum and taxane exposure. Therefore, in this section, we present data from relevant phase II and phase III trials with respect to these important prognosticators (particularly platinum sensitivity and treatment-free survival) in order to provide some relative parity for context. The options of salvage chemotherapy for recurrent ovarian cancer include treatment with (1) single-agent platinum compound, (2) single-agent nonplatinum drug,
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
(3) platinum-based combination regimen, and (4) nonplatinum-based combination regimen. Single-agent platinum Since the platinum compounds are generally regarded as the most active agents in advanced ovarian cancer, platinum-based therapy is often considered first when platinum-sensitive disease recurs. Patients who have a prolonged duration of response to first-line platinum-based therapy have an increased likelihood of response to platinum retreatment [13,14]. Combined data from the landmark Gore and Markman studies show increased response rates with increasing TFI up to 24 months and beyond; in fact, the response rate to a platinum agent after a TFI of 5–12 months is 22%; compared to 31% and 59%, respectively, for TFI of 12–24 and N24 months. These trials were largely conducted in an era of relatively few effective chemotherapeutic alternatives for women with recurrent ovarian cancer. Nonetheless, a study confirming these observations was recently presented by Markman et al. who retreated 214 platinum-sensitive patients with predominately single-agent carboplatin. Responses were seen overall in 55% of patients but this was significantly influenced by TFI. Those patients with a TFI of less than 12 months had a 35% response rate; those with TFI greater than 36 months had a response rate of 80%. Single-agent nonplatinum Similar bTFI-relatedQ effects are also seen with many of the novel nonplatinum agents. For instance, in the large randomized trial of recurrent ovarian cancer patients treated with either single-agent liposomal doxorubicin or topotecan, clinical response rates were 12% and 7%, respectively, in 254 platinum-resistant patients (defined as progression on or recurrence within 6 months of front-line therapy), and 28% and 29%, respectively among 220 platinum-sensitive patients [15]. Survival was similarly influenced in this trial, being a median 36 and 41 weeks, respectively, in platinumresistant patients treated with liposomal doxorubicin or topotecan, and 108 and 71 weeks in platinum-sensitive patients. Though debatable, some investigators have reported that progression-free survival may be the most important factor in determining subsequent tumor response and survival. In a review of eight GINECO Phase II and III chemotherapy clinical trials involving 583 patients in the recurrent setting, Pujade-Lauraine et al. [16] demonstrated that response and survival were closely related to initial relapse-free interval. Patients recurring in 3–12 months of their initial treatment had a response rate of 35% and an overall survival of 393 days; in the cohort with a relapse-free survival of at least 18 months, response was 62% and overall survival was 957 days. Importantly, these investigators found these effects to be independent of the individual chemotherapeutic agents.
277
This positive influence of a longer TFI on patients’ survival is similar to what has been described in the secondary cytoreduction studies. Thus, it would be reasonable to expect that patients generally considered for secondary surgery followed by chemotherapy would enjoy by selection, a longer survivorship than those with shorter TFI treated with chemotherapy alone. Combination regimens Another popular treatment strategy in patients with long TFI is the use of combination chemotherapy. Some clinicians maintain that a combination regimen may be better than single-agent carboplatin or cisplatin, since some phase II data suggest somewhat higher response rates and disease-free survivals with combination therapy. Markman et al. [17] made the observation that in their patients’ secondary TFI were temporally shorter than the initial TFI; however, the reverse was seen in three patients who received combination salvage chemotherapy suggesting a beneficial role for combination chemotherapy in the salvage setting. Dizon et al. [18] reported a 70% overall response rate (42% CR) among 66 drug-sensitive patients (median treatmentfree interval 22 months) with combination paclitaxel and carboplatin. The median progression-free interval following therapy was 13 months and 3-year overall survival was 72%. In order to place these data into perspective against a single-agent approach, these same authors recently reported data on a similar cohort treated with single-agent carboplatin [19]. All response parameters were significantly lower among the 22 patients undergoing single-agent therapy including response rates (36% vs. 70%) and median survival (26 vs. 42 months). The median treatment-free interval was 22 months. These favorable responses must be carefully interpreted against their study cohort, which was significantly heterogeneous. It is clear that combination therapy in platinum-sensitive patients is being increasingly evaluated in phase II clinical trials. Most are incorporating agents that have demonstrated historical activity in ovarian cancer or have the potential to demonstrate additive or synergistic activity. One trial highlighting this latter point was reported by Nagourney et al. [20] studying the gemcitabine/cisplatin combination in a mixed group of recurrent ovarian cancer patients (n = 27). Evidence for synergistic activity between these agents is well documented and appears to be related not only to cisplatin’s ability to inhibit ribonucleotide reductase, thereby reducing the rate-limiting inhibition of gemcitabine phosphorylation, but also through enhanced inhibition of the repair of cisplatin-induced DNA adducts. In this small trial, 11 of 13 (84%) patients with a median platinum-free interval of 12 months responded including four complete responders. Overall survival for the study population was 20.9 months. Prospective data comparing single-agent to combination chemotherapy are very limited (Table 2). Bolis et al. [21] reported on 190 platinum-sensitive (N6 months) patients who
278
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
Table 2 Comparative studies on salvage chemotherapy Agent/combination chemotherapy [references]
N
% platinumsensitivea
TFI (months)
PFS (weeks)
P
OS (weeks)
P
Topotecan vs. paclitaxel [24] Paclitaxel vs. weekly paclitaxel [25] Oxaliplatin vs. paclitaxel [26] DoxilR vs. topotecan [15] Topotecan vs. treosulfan [27] Paclitaxel vs. paclitaxel/epirubicin [28] Carboplatin vs. carboplatin/epirubicin [21] Paclitaxel vs. CAP [29] Platinum regimen vs. platinum/paclitaxel [22] Carboplatin vs. carboplatin/gemcitabine [23]
226 208 86 481 357 81 190 97 802 356
47 48 26 46 Mixed – – – – –
– – – – – 1.5 17 34.5 N12 N6
23 38 12 16 22 26 65 39 43 25
NS NS NS NS 0.001 NS NS 0.04 b0.001 0.004
61 vs. 64 vs. 42 vs. 60 vs. 56 vs. 39 vs. 109 vs. 112 vs. 104 vs. –
NS NS NS NS 0.02 NS NS 0.04 0.023 –
vs. vs. vs. vs. vs. vs. vs. vs. vs. vs.
14 26 14 17 12 43 78 68 52 37
43 59 37 57 48 61 122 151 130
TFI: treatment-free interval; PFS: progression-free survival; OS: overall survival. a Percent of the platinum-sensitive patients in the study population.
were randomized to receive either single agent carboplatin (300 mg/m2) or combination carboplatin (300 mg/m2) and epirubicin (120 mg/m2). The median platinum-free interval was 17 months and a minimum of five cycles of therapy was planned. The overall response rate was 55% for carboplatin vs. 58% for the combination. Median progression-free and overall survivals favored the combination regimen but were not statistically significant being 15 and 25 months in the carboplatin arm and 18 and 28 months in the carboplatin/ epirubicin arm, respectively. Limitation in sample size may have contributed to the effect. Toxicity, however, was significantly increased, with hematological and gastrointestinal toxicities occurring more frequently in the combination cohort. Recently, the ICON collaborators conducted a large randomized trial of paclitaxel and platinum vs. platinumbased conventional therapy in first relapse platinum-sensitive patients [22]. This 802 patient trial represents the largest such trial completed to date in this setting. Patients were allowed entry following primary platinum-based chemotherapy and were progression-free a minimum of 6 months in patients recruited from the Medical Council’s Clinical Trials Unit (MRC) and Arbeitsgemeinschaft Gynaekologische Onkologie (AGO) and a minimum 12 months in those recruited from the Instituto Mario Negri (IRFMN). Overall response was 66% for the combination and 54% for the single agent ( P = 0.06). A significant improvement in progression-free (9 vs. 12 months, HR = 0.76, P b 0.001) and overall survival (24 vs. 29 months, HR = 0.82, P = 0.023) were observed for the combination of paclitaxel and platinum at a median follow-up of 42 months. Toxicity was mixed between the arms with more hematologic toxicity seen in the conventional platinum arm particularly singleagent carboplatin, and more neurotoxicity was observed in the taxane combination regiments. This provocative data, if confirmed, will likely justify combination paclitaxel and platinum chemotherapy in the cohort of patients also strongly considered for secondary surgery, and may be the most promising strategy following a salvage surgical intervention. A second prospective randomized trial conducted by the Gynecologic Cancer Intergroup (GCIG) has
been recently reported [23]. In this trial, 356 patients with recurrent ovarian cancer were allocated to either single agent carboplatin or combination carboplatin and gemcitabine. Currently, data on the time to progression endpoint confirm an improved activity for the combination regimen. In the trial, the hazard for progression after treatment was 0.72 (95% CI: 0.6–0.9) amounting to a 12-week increase in progression-free interval (23 vs. 35 wks). Response rates were also higher in the combination (47% vs. 31%, P = 0.0016). As expected toxicity, predominately hematologic, was more common in the combination arm. The unique combination, which may highlight intrinsic drug synergy, illustrates that other combinations may also be of interest in this setting.
Conclusion Although marginally less biased than the aforementioned surgical trials, the collective experience from the chemotherapy studies would suggest that equally selected platinum-sensitive patients undergoing salvage chemotherapy with no surgery may be expected to have durable survival. In the overall experience from the presented phase II, randomized single-agent and randomized combination chemotherapy trials, overall survival ranged between 24 and 42 months. However, none of the chemotherapy trials reports median survivals that are as prolonged as those seen in patients who undergo complete resection of tumor recurrence and receive adjuvant salvage chemotherapy. One can argue that a longer progression-free interval after primary therapy reflects a tumor that is biologically less aggressive and more chemoresponsive. Therefore, patients who have recurrent disease after a prolonged progressionfree interval might benefit most from a combined approach at complete tumor cytoreduction and chemotherapy similar to the approach adopted in primary treatment. It is clear that the available literature does not offer the answer as to how patients should be selected for secondary cytoreduction, who would benefit from such surgical
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
279
Fig. 1. Proposed schema for protocol 213.
approach, and what impact this will have on their quality of life; this will only be determined from carefully controlled randomized trials. Given the nature of such trials and the number of prognostic factors such as patient age and performance status, disease stage, length of progressionfree interval, and extent of recurrent disease, these studies will be necessarily large and ambitious. In addition, the recent data suggesting the added survival benefit of salvage combination chemotherapy necessitate a careful analysis of the types of chemotherapy regimens delivered after surgery. Since the benefit of the latter has only been demonstrated in one randomized trial, balancing the effect of two aggressive approaches, surgery and combination chemotherapy, with quality of life will be important in order to provide the context with which to interpret any observed benefit. Currently, one such trial from the EORTC (Protocol# 55963) is underway and accruing patients. The inclusion criteria from this trial are disease recurrence, after a minimum of three cycles of primary chemotherapy and—a disease-free interval z12 months. Patients are randomized to either six cycles of single-agent platinum (carboplatin or cisplatin) or three cycles of the same chemotherapy, followed by surgical exploration and three more cycles of chemotherapy. They are stratified by initial stage, progression-free survival (b24 vs. N24 months), initial response to primary therapy, and tumor burden. Anticipated accrual is 700 patients and survival is the primary endpoint. A second trial, shown in Fig. 1, is being conducted through the GOG (Protocol #213). This trial randomizes recurrent ovarian and peritoneal patients with a minimum treatment-free interval of 6 months to either surgical exploration or chemotherapy. Given the provocative data presented above, a secondary randomization is being
made between platinum-based combination chemotherapy, combination of chemotheraphy and biologics and a control arm defined as single agent carboplatin followed by a paclitaxel in third line. It is hoped that those trials, when completed, will provide more conclusive data as to the optimal management of patients who suffer from recurrent ovarian cancer after a complete remission.
References [1] Morris M, Gershenson M, Wharton T, Copeland LJ, Edwards CL, Stringer CA. Secondary cytoreductive surgery for recurrent epithelial ovarian cancer. Gynecol Oncol 1989;34:334 – 8. [2] Janicke F, Holscher M, Kuhn W, Hugo RV, Pache L, Siewert JR, et al. Radical surgical procedure improves survival time in patients with recurrent ovarian cancer. Cancer 1992;70. [3] Vaccarello L, Rubin SC, Vlamis VGW, Jones WB, Lewis JL, Hoskins WJ. Cytoreductive surgery in ovarian carcinoma patients with a documented previously complete surgical response. Gynecol Oncol 1995;57:61 – 5. [4] Eisenkop SM, Friedman RL, Spirtos NM. The role of secondary cytoreductive surgery in the treatment of patients with recurrent epithelial ovarian carcinoma. Cancer 2000;88. [5] Gadducci A, Iacconi P, Cosio S, Fanucchi A, Cristofani R, Genazzani AR. Complete salvage surgical cytoreduction improves further survival of patients with late recurrent ovarian cancer. Gynecol Oncol 2000;79:344 – 9. [6] Zang RY, Li ZT, Tang J, Cheng X, Cai SM, Zhang ZY, et al. Secondary cytoreductive surgery for patients with relapsed epithelial ovarian carcinoma: who benefits? Cancer 2004;100:1152 – 61. [7] Munkarah AR, Levenback C, Wolf JK, Bodurka-Bevers D, TortoleroLuna G, Morris R, et al. Secondary cytoreductive surgery for localized intra-abdominal recurrences in epithelial ovarian cancer. Gynecol Oncol 2001;81:237 – 41.
280
A.R. Munkarah, R.L. Coleman / Gynecologic Oncology 95 (2004) 273–280
[8] Scarabelli C, Gallo A, Carbone A. Secondary cytoreductive surgery for patients with recurrent epithelial ovarian carcinoma. Gynecol Oncol 2001;83:504 – 12. [9] Tay EH, Grant PT, Gebski V, Hacker NF. Secondary cytoreductive surgery for recurrent epithelial ovarian cancer. Obstet Gynecol 2002;99. [10] Merideth MA, Cliby WA, Keeney GL, Lesnick TG, Nagorney DM, Podratz KC. Hepatic resection for metachronous metastases from ovarian carcinoma. Gynecol Oncol 2003. [11] Yoon SS, Jarnagin WR, Fong Y, DeMatteo RP, Barakat RR, Blumgart LH, et al. Resection of recurrent ovarian or fallopian tube carcinoma involving the liver. Gynecol Oncol 2003;91:383 – 8. [12] Zang RY, Zhang ZY, Li ZT, Chen J, Tang MQ, Liu Q, et al. Effect of cytoreductive surgery on survival of patients with recurrent epithelial ovarian cancer. J Surg Oncol 2000;75:24 – 30. [13] Markman M, Rothman R, Hakes T. Second-line platinum therapy in patients with ovarian cancer previously treated with cisplatin. J Clin Oncol 1991;9:389 – 93. [14] Gore ME, Fryatt I, Wiltshaw E. Cisplatin/carboplatin cross-resistance in ovarian cancer. Br J Cancer 1989;60:767 – 9. [15] Gordon AN, Fleagle JT, Guthrie D. Recurrent epithelial ovarian carcinoma: a randomized phase II study of pegylated liposomal doxorubicin versus topotecan. J Clin Oncol 2001;19:3312 – 22. [16] Pujade-Lauraine E, Paraiso D, Cure H. Predicting the effectiveness of chemotherapy (CX) in patients with recurrent ovarian cancer (ROC): a GINECO study: Abst#829. Proc Am Soc Clin Oncol 2002. [17] Markman M, Markman JR, Zanotti KM. Duration of response to second-line platinum-based chemotherapy for ovarian cancer: implication for patient management and clinical trial design. Proc Am Soc Clin Oncol 2003;22:447. [18] Dizon DS, Hensley ML, Poynor EA. Retrospective analysis of carboplatin and paclitaxel as initial second-line therapy for recurrent epithelial ovarian carcinoma: application toward a dynamic disease state of model of ovarian cancer. J Clin Oncol 2002;20:1238 – 47. [19] Dizon DS, Dupont J, Anderson S. Treatment of recurrent ovarian cancer: a retrospective analysis of women treated with single-agent carboplatin originally treated with carboplatin and paclitaxel. The Memorial Sloan-Kettering Cancer Center experience. Gynecol Oncol 2003;91:584 – 90.
[20] Nagourney RA, Brewer CA, Radecki S. Phase II trial of gemcitabine plus cisplatin repeating doublet therapy in previously treated, relapsed ovarian cancer patients. Gynecol Oncol 2003;88:35 – 9. [21] Bolis G, Scarfone G, Giardina G. Carboplatin alone vs. carboplatin plus epidoxorubicin as second-line therapy for cisplatin- or carboplatin-sensitive ovarian cancer. Gynecol Oncol 2001;81:3 – 9. [22] Parmar MK, Ledermann JA, Colombo N. Paclitaxel plus platinumbased chemotherapy versus conventional platinum-based chemotherapy in women with relapsed ovarian cancer: the ICON4/AGOOVAR-2.2 trial. Lancet 2003;361:2099 – 106. [23] Pfisterer J, Plante M, Vergote I. Gemcitabine/carboplatin vs. carboplatin in platinum sensitive recurrent ovarian cancer. Results of a Gynecologic Cancer Intergroup randomized phase III trial of the AGO OVAR, the NCIC CTG, and the EORTC GCG. Proc Am Soc Clin Oncol 2004;23:449. [24] ten Bokkel Huinink W, Gore M, Carmichael J. Topotecan versus paclitaxel for the treatment of recurrent epithelial ovarian cancer. J Clin Oncol 1997;15:2183 – 93. [25] Rosenberg P, Anderson H, Boman K. Randomized trial of single agent paclitaxel given weekly versus every three weeks and with peroral versus intravenous steroid premedication to patients with ovarian cancer previously treated with platinum. Acta Oncol 2002;41:418 – 24. [26] Piccart MJ, Green JA, Lacave AJ. Oxaliplatin or paclitaxel in patients with platinum-pretreated advanced ovarian cancer: a randomized phase II study of the European Organization for Research and Treatment of Cancer Gynecology Group. J Clin Oncol 2000;18: 1193 – 202. [27] Meier W, du Bois A, Kuhn W. Topotecan vs. treosulfan in recurrent ovarian cancer after initial therapy with platinum and paclitaxel. A prospective randomized phase III study. Proc Am Soc Clin Oncol 2003;22:450. [28] Bolis G, Parazzini F, Scarfone G. Paclitaxel vs. epidoxorubicin plus paclitaxel as second-line therapy for platinum-refractory and-resistant ovarian cancer. Gynecol Oncol 1999;72:60 – 4. [29] Cantu MG, Buda A, Parma G, Rossi R, Floriani I, Bonazzi C, et al. Randomized controlled trial of single-agent paclitaxel versus cyclophosphamide, doxorubicin, and cisplatin in patients with recurrent ovarian cancer who responded to first-line platinum-based regimens. J Clin Oncol 2002;5:1158 – 60.