- Email: [email protected]
Impact of bevacizumab containing first line chemotherapy on recurrent disease in epithelial ovarian cancer: A case-control study
Gynecologic Oncology xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Society Position Statements/White Papers
Impact of bevacizumab containing first line chemotherapy on recurrent disease in epithelial ovarian cancer: A case-control study M. Petrillo a, G. Amadio a, V. Salutari a, I. Paris a, M.G. Di Stefano a, G. Ferandina b,c, G. Scambia a, A. Fagotti d,e,⁎ a
Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Catholic University of the Sacred Heart, Rome, Italy Department of Medicine and Health Sciences, University of Molise, Campobasso, Italy Gynecologic Oncology Unit, Foundation “Policlinico Universitario A. Gemelli”, Rome, Italy d Department of Obstetrics and Gynecology, Catholic University of the Sacred Heart, Rome, Italy e University of Perugia, Italy b c
H I G H L I G H T S • Incorporation of bevacizumab into upfront regimens prolongs PFI in AOC patients. • However, it is associated with wider presentation of relapse, and lower rate of complete SCS. • TTP to second line chemotherapy was shorter in women with platinum-sensitive relapse initially treated with bevacizumab.
a r t i c l e
i n f o
Article history: Received 2 April 2016 Received in revised form 7 May 2016 Accepted 16 May 2016 Available online xxxx Keywords: Bevacizumab Recurrent ovarian cancer Secondary cytoreductive surgery Response to second-line chemotherapy
a b s t r a c t Objective. To evaluate the timing and pattern of relapse, and duration of response to second line chemotherapy in advanced ovarian cancer (AOC) patients treated with first line carboplatin-paclitaxel chemotherapy with or without bevacizumab. Patients and methods. This is a case-control study including 222 AOC patients. Seventy-four women treated with first line carboplatin-paclitaxel-bevacizumab chemotherapy (Cases) were matched based on laparoscopic predictive index value, and residual tumor at first surgery with 148 AOC patients treated with carboplatinpaclitaxel. Distribution of pattern of relapse, and response to second line chemotherapy was compared between the two groups. Time to Progression (TTP) for second line chemotherapy was also analyzed for study purpose. Results. Median platinum-free interval (PFI) was 16 months (range 2–65) in Cases, compared with 9 months (1–83) in Controls (p-value = 0.001). Twenty patients (51.3%) among Cases showed recurrence in multiple anatomic sites, compared with 31 (31.9%) in the Control group (p-value = 0.035). Peritoneal recurrence occurred as diffuse in 30 Cases (96.8%), and 60 Controls (82.2%; p-value = 0.046). Secondary cytoreductive surgery (SCS) was successfully completed in 53.5% of Controls compared to 10.0% of Cases (p-value = 0.016). In women with fully platinum-sensitive relapse, response rate to second line chemotherapy was 85.2% in Controls, compared to 38.4% in Cases (p-value = 0.002). Finally, Cases showed a shorter TTP, compared to Controls (5 months vs 8 months; p-value = 0.041). Conclusions. Incorporation of bevacizumab into upfront regimens prolongs PFI in AOC patients, but is associated with wider presentation of relapse, lower rate of complete SCS, and shorter TTP to second line chemotherapy in women with platinum-sensitive disease. © 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction In the past decade, the results of two pivotal randomized phase III clinical trials demonstrated a significant improvement of progression⁎ Corresponding author at: Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Catholic University of the Sacred Heart, Largo A. Gemelli, 8, 00168 Rome, Italy. E-mail address: [email protected] (A. Fagotti).
free survival (PFS), in patients with advanced ovarian cancer (AOC) receiving bevacizumab as part of first line treatment [1,2]. Based on these data, in December 2011, Avastin received European regulatory approval for use in combination with carboplatin-paclitaxel as upfront chemotherapy regimen in OC patients with advanced disease. The observed prolongation of PFS should be considered a relevant therapeutic achievement, since it delays the onset of physical symptoms associated with progressive disease, as well as the need to start a second-line chemotherapy. Furthermore, the prolongation of PFS may
http://dx.doi.org/10.1016/j.ygyno.2016.05.017 0090-8258/© 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
2
M. Petrillo et al. / Gynecologic Oncology xxx (2016) xxx–xxx
in principle increase the chance of response to further platinum-based treatments [3], thus offering a potential long-term benefit. However, in this encouraging scenario, it is reasonable to hypothesize that the incorporation of bevacizumab into first-line treatment may exert an additional selective pressure on OC cells, with a potentially relevant impact on the natural history of this malignancy. This hypothesis is also supported by recently published retrospective data, which demonstrated a higher rate of lung and pleural recurrence in AOC patients receiving bevacizumab as part of primary treatment [4], thus suggesting that more efforts should be done to definitely clarify how bevacizumab will change the clinical features of OC. Furthermore, despite initial promising results [5,6], it remains to be validated an effective signature able to properly identify at diagnosis those AOC patients who may real benefit from first line bevacizumab-based chemotherapy. For these reasons, we focused our attention on the impact of upfront bevacizumab containing regimens on recurrent disease, analyzing the timing and pattern of relapse, as well as the rate and duration of response to second line chemotherapy in a single-Institutional case control study including AOC patients treated with first line carboplatinpaclitaxel chemotherapy with or without bevacizumab. 2. Patients and methods 2.1. Study groups The investigational arm (Cases) included 74 patients with AOC admitted at the Gynaecologic Oncology Unit of the Catholic University of Rome and Campobasso between March 2010 and February 2013 treated, after joint evaluation with the oncologists' team, with the following regimen: carboplatin 5 AUC/paclitaxel 175 mg per square meter of body-surface area/bevacizumab 15 mg/Kg every three weeks, followed by maintenance therapy with bevacizumab 15 mg/Kg until disease progression or for a maximum of 15 months. The Control arm included a consecutive series of 148 AOC patients, admitted at our Institution between January 2010 and December 2012, treated with carboplatin 5 AUC/paclitaxel 175 mg per square meter of body-surface area every three weeks. In both groups, patients were selected for primary debulking surgery (PDS), or neoadjuvant chemotherapy based on our Institutional laparoscopy-based algorithm [7], and to avoid imbalance between the two groups in term of initial disease extension and clinical features the Cases were matched with the Controls according to the laparoscopic predictive index value [7,8], and residual tumor at first surgery, using Stata software version 11.0 (Stata Corp, College station, TX). Furthermore, to maximize the power of the study, Cases and Controls were matched in a 1:2 ratio. All women gave a written informed consent for their data to be collected and analyzed for scientific purpose. The Institutional Review Board approved the study.
All patients showing platinum-sensitive recurrent disease were submitted to FDG-PET/CT scan, and staging-laparoscopy, to assess the chance of optimal secondary cytoreduction (SCS) [10,11]. All patients judged as suitable for optimal SCS received an attempt of surgical debulking. Patients showing platinum-sensitive recurrent disease received second-line platinum-based chemotherapy; on the other hand, women showing platinum-resistant relapse were treated with nonplatinum agents including weekly paclitaxel, pegylated liposomal doxorubicin, and gemcitabine. The group of cases developing recurrence with a PFI between 6, and 12 months were triaged to receive platinum versus non-platinum regimens, based on a case-specific decision making process. Clinical response to second line chemotherapy was assigned according to RECIST criteria [12], and the response has been classified as complete (CR)/partial (PR), or stable (SD)/progressive (PD). In order to fully evaluate the efficacy of second line treatment, we analyzed in our study also the Time to Progression (TTP), which was defined as the time from beginning of second line chemotherapy to further progressive disease. Data analysis regarding overall survival was not attempted due to the duration of follow-up period. 2.3. Statistical analysis Differences between Cases and Controls in term of clinicopathological features at diagnosis, and at the time of relapse were analyzed using the Pearson Chi-square exact test and Kruskall-Wallis test, as appropriate. Medians and life tables were computed using the product limit estimate by Kaplan–Meier method [13], and the log-rank test was used to assess the statistical significance [14]. Cox's regression model with stepwise variable selection [15] was used to analyse the role of clinical-pathological parameters, and treatment details as predictors of TTP for second line chemotherapy. All statistical calculations were performed using the Stata software version 11.0 (Stata Corp, College station, TX). 3. Results 3.1. Clinico-pathological features at diagnosis As a consequence of matching process, no differences were observed between Cases and Controls in term of disease extension at diagnosis evaluated as laparoscopic predictive index value, and residual tumor at first surgery (Table 1). Around 90% of AOC patients showed highgrade serous histology, and FIGO Stage IIIC at initial diagnosis, without differences between the two groups (Table 1). Median CA125 levels were 792 IU (range 10–7654) in the overall population, with similar levels in Cases and Controls (p-value = 0.607). Median age at diagnosis of women treated with first line bevacizumab-containing chemotherapy was 54 years (range 26–76), which appears slightly lower compared with Controls (median 59, range 29–86; p-value = 0.069).
2.2. Clinical data
3.2. Timing of recurrent disease
For all patients included in the final analysis, data regarding timing, pattern, and treatment details at the time of relapse were prospectively recorded, and retrospectively analyzed for study purpose. In particular, the timing of recurrent disease was evaluated using the PFI defined as the time elapsed between the date of completion of platinum-based first line chemotherapy and recurrence. The pattern of relapse was classified as peritoneal, parenchymal, and lymph-nodal based on the site of disease [9]. Furthermore, we defined the type of recurrence as single in presence of only one anatomic site involved, or multiple when relapse diffusion occurred in more than one site. Finally, the extension of intraperitoneal recurrence was classified as localized in presence of no more than three nodules, or diffused when a wider disease spread was observed [9].
After a median follow-up of 32 months (Cases 25 months, Controls 32 months; p-value = 0.785), we observed recurrent disease in 39 Cases (52.7%), and 116 Controls (78.4%; p-value = 0.001; Fig. 1). Median PFI was 16 months (range 2–65) in Cases, compared with 9 months (4–83) in the Control group (p-value = 0.001; Fig. 2). AOC patients initially treated with PDS showed a median PFI of 10 months (1–83) in Controls, compared with 19 months (2–65) in Cases (p-value = 0.001, Table 2). Similarly, in the group of women treated with upfront NACT followed by interval debulking surgery, we observed a longer PFI in patients treated with bevacizumab (10 months) compared with Controls (7 months; p-value = 0.032; Table 2). The distribution of relapse according with PFI in the two groups has been reported in Table 2. In particular, we observed platinum-resistant relapse in 45
M. Petrillo et al. / Gynecologic Oncology xxx (2016) xxx–xxx
3
Controls n (%)
p valuea
74 (33.3) 54 (26–76)
148 (66.7) 59 (29–86)
0.069
202 (90.9) 11 (4.9) 9 (4.2)
64 (86.4) 5 (6.8) 5 (6.8)
138 (93.2) 6 (4.1) 4 (2.7)
0.134
only in the subgroup of women with platinum-sensitive recurrent disease. Furthermore, as presented in Table 3, TTP was analyzed according to initial treatment strategy (NACT or PDS), and in the group of women who received PDS, duration of TTP was 6 months (1–24) in Cases, and 9 months (1–39) in Controls (p-value = 0.175). Similarly, also in patients initially selected for NACT, we observed a longer median TTP in Controls (6 months) compared with Cases (4 months, p-value = 0.023; Table 3). Finally, at multivariate analysis, the administration of bevacizumab in the context of first line treatment confirmed its independent negative prognostic role for duration of TTP (p value = 0.047; Supplementary Table 1).
85 (38.3) 51 (20.0) 86 (38.7)
28 (37.8) 17 (23.0) 29 (39.2)
57 (38.5) 34 (23.0) 57 (38.5)
0.994
4. Discussion
203 (91.4) 19 (8.6)
67 (90.5) 7 (9.5)
136 (91.9) 12 (8.1)
0.800
Table 1 Distribution of patients' clinico-pathological characteristics at diagnosis in cases and controls. Characteristics
All pts
All Age (median and range, years)b Tumor histotypes High-grade serous Endometrioid Clear cell LPS-PIV at diagnosis b4 4–6 ≥8 FIGO stage IIIC IV Residual tumor at first surgery Complete ≤1 cm NACT CA125 (median and range, UI/ml)b
222 58 (26–86)
Cases n (%)
132 (59.5) 44 (59.5) 88 (59.5) 20 (9.0) 6 (8.1) 14 (9.5) 70 (31.5) 24 (32.4) 46 (31.1) 0.938 792 (10–7654) 707 (10–9842) 852 (24–7654) 0.667
LPS-PIV = laparoscopic predictive index value; NACT = neoadjuvant chemotherapy. a Calculated by Chi-square test. b Calculated by Kruskal-Wallis non-parametric test.
patients (29.0%), partially platinum-sensitive recurrent in 63 women (40.6%), and fully platinum-sensitive disease in 47 (30.4%) cases, without differences between the two groups (p-value = 0.636; Table 2). 3.3. Pattern of recurrent disease Focusing on pattern of disease presentation, 20 patients (51.3%) among Cases showed involvement of multiple anatomic sites, compared with 31 (31.9%) women in the Control group (p-value = 0.035). This wider recurrent spread was sustained by higher incidence of peritoneal and lymph nodal relapse in Cases compared with Controls (Table 2). Finally, peritoneal recurrence occurred more frequently as diffuse disease in patients treated with bevacizumab-based first line regimen (30 patients, 96.8%), compared with women receiving conventional treatment (60 patients, 82.2%; p-value = 0.046). 3.4. SCS and response to second line chemotherapy As a consequence of differences in term of disease spread, SCS was successfully completed in a larger group of Controls (53.5%) compared to Cases (10.0%, p-value = 0.016; Table 3). Overall, platinum-based second line chemotherapy was administered in 78 patients (50.3%), while 77 (49.7%) women received non-platinum containing regimens, without differences between the two groups (p-value = 0.481). Response to second line chemotherapy (CR + PR) was observed in 9 Cases (23.1%), and 47 Controls (40.5%), with a trend toward a statistically significant difference (p-value = 0.055; Table 3). Furthermore, when response rate was analyzed according with duration of PFI, we documented a statistically significant difference in favour of Controls (85.2%) compared to Cases (38.4%, p-value = 0.002; Table 3) in the subgroup of women with fully platinum-sensitive relapse. 3.5. Time to progression for second line chemotherapy At Kaplan-Meier analysis, Cases showed a shorter median TTP, compared to Controls (5 months vs 8 months; p-value = 0.041; Fig. 3). However, when duration of TTP was analyzed according with PFI, median TTP resulted significantly longer in Controls compared to Cases (5 months vs 10 months; p-value = 0.006; Supplementary Fig. 1),
The observation of a longer PFS in women with AOC receiving bevacizumab-containing first line chemotherapy certainly represents a relevant achievement in the management of epithelial OC; furthermore, a survival benefit related to bevacizumab administration was demonstrated also in other clinical settings including women with platinumsensitive [16,17], and resistant recurrent disease [18,19]. However, even if the overall efficacy of bevacizumab appears unquestionable, several issues are still open, including: the best clinical setting in which the drug should be administered to maximize overall survival [20], the most appropriate duration of maintenance therapy (AGO-OVAR17; NCT01462890; ROSiA NCT01239732), and whether bevacizumab administration could give a benefit beyond progression (MITO-16/ MaNGO OV-2, NCT01802749). In this continuously changing horizon, we present here, the first single Institution case-control study comparing the features of recurrent disease in a large series of AOC patients treated with or without bevacizumab as first line treatment. The accurate matching in term of disease extension, using as measure the laparoscopic index value, and the exact balancing of cases and controls in term of residual tumor at first surgery represents major strengthens of our study, thus ultimately supporting the reliability of our findings. For instance, we confirmed in our series that the incorporation of bevacizumab into first line treatment ensures an around seven months prolongation of PFI, in the overall population of AOC patients [1–2], and we confirmed previously published data suggesting a bevacizumabrelated benefit in term of PFS also in patients treated with NACT [21]. However, despite the longer PFS in the group of women treated with bevacizumab, we did not observe differences in the distribution of patients with fully-sensitive, partially-sensitive or platinum-resistant relapse between the Cases and Controls. Based on these data, it may be hypothesized that the major contribution of bevacizumab upfront administration is mainly represented by an overall reduction of recurrence rate, rather than a prolongation of PFI. On the other hand, as previously reported in a retrospective analysis [21], we documented differences in term of pattern of disease presentation between the two groups. In particular, we observed a more frequent involvement of multiple anatomic sites in relapsed AOC patients initially treated with bevacizumab-based regimens. Furthermore, focusing on intraperitoneal recurrence we documented a statistically significant higher incidence of diffuse disease in Cases compared with Controls. Far from drawing definitive conclusions regarding the survival impact of the more unfavourable pattern of relapse in AOC patients initially treated with bevacizumab, our data may contribute to explain the lack of a clear benefit in term of overall survival emerged in Phase III randomized clinical trials [1–2]. Indeed, the observed disease diffusion at the time of recurrence carries on relevant therapeutic implications. In particular, as a consequence of the more frequent involvement of multiple anatomic sites, and of the wider diffusion of peritoneal disease, SCS was successfully attempted in only 10% of women previously treated with bevacizumab, compared with around 50% of women receiving optimal SCS in the Control group.
4
M. Petrillo et al. / Gynecologic Oncology xxx (2016) xxx–xxx
Fig. 1. Flow-chart of our study population.
Another relevant result of our analysis is represented by trend toward a lower response rate, and a three months shorter statistically significant TTP to second line chemotherapy in favour of Controls compared with Cases. The differences in term of response rate appear of higher magnitude, reaching statistical significance, in patients with fully-platinum sensitive recurrent disease. Similarly, the longer TTP to second line chemotherapy in favour of Controls documented in the overall population was not observed in the group of platinum-resistant disease, being however confirmed in patients with a PFI ≥ 6 months. It could be argued that the above mentioned differences between Cases and Controls in term of response rate, ant TTP to second line chemotherapy may be the consequence of wider pattern of disease presentation, and a lower percentage of SCS in the group of Cases. However, taken together our findings, and previously published experiences [4], suggest that the incorporation of bevacizumab into first line regimens, even ensuring an overall PFS benefit, influences the disease natural history, and in particular the clinical features, and management of recurrent disease.
In this context, recently published data demonstrated in preclinical models a potential role of tumor-associated macrophages in driving resistance to antiangiogenic agents [22–23], thus emphasizing that further biological treatment strategies should be developed to reduce the clinical impact of the mechanisms involved in determining the acquisition of adaptive resistance to antiangiogenic agents [24]. In conclusions, despite the exact matching in term of disease extension, and residual tumor at first surgery, the retrospective nature, and the small sample size certainly represent relevant study limitations. However, awaiting results from randomized clinical trials, our data suggest for the first time, in the context of a case-control study, that the incorporation of bevacizumab into upfront regimens prolongs PFI in AOC patients, but is associated with a more aggressive behaviour of recurrent disease, including a lower percentage of patients suitable for SCS, and a shorter TTP to second line chemotherapy in women with platinum-sensitive disease. Larger studies are needed to confirm these data, and a multi-Institutional commitment is urgently required to identify molecular signatures able to predict response to bevacizumab first line treatment, and to clarify the biological effects of chronic antiangiogenic exposure of OC cells.
M. Petrillo et al. / Gynecologic Oncology xxx (2016) xxx–xxx
5
Fig. 2. Platinum-free interval in Cases (solid line) and Controls (dotted line).
Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ygyno.2016.05.017.
Conflict of interest statement The authors declare that there are no conflicts of interest.
Table 2 Timing and pattern of recurrent disease in women receiving or not bevacizumab containing first line chemotherapy. Characteristics
All cases n (%)
Recurrence PFIb (median and range, months) PFIb (median and range, months) Primary debulking surgery +
155 (69.8) 39 (52.7) 116 (78.4) 0.001 11 (1–83) 16 (2–65) 9 (1–83) 0.001
adjuvant chemotherapy NACT followed by IDS Timing of relapse (months) PFI b 6 months 6 ≤ PFI b 12 months PFI ≥ 12 months Site of recurrence Parenchyma (lung, spleen, liver) Lymph nodes Peritoneum Type of recurrence Single site Multiple sites Extension of intraperitoneal recurrence Localized Diffused
Cases n (%)
Controls n (%)
p valuea
14 (1–83) 19 (2–65)
10 (1–83) 0.001
8 (1–36) 10 (1–36)
7 (1–18) 0.032
45 (29.0) 63 (40.6) 47 (30.4)
9 (23.1) 17 (43.6) 13 (33.3)
36 (31.0) 46 (39.7) 34 (29.3)
18 (11.4) 50 (32.1) 104 (66.5)
6 (15.4) 20 (51.3) 31 (79.5)
12 (10.3) 30 (25.9) 73 (62.9)
98 (63.2) 57 (36.8)
19 (48.7) 20 (51.3)
79 (68.1) 37 (31.9)
17 (10.9) 138 (89.1)
1 (3.2) 30 (96.8)
13 (17.8) 60 (82.2)
0.636 0.279 0.004 0.041
0.035
0.046
PFI = platinum-free interval; NACT = neoadjuvant chemotherapy; IDS = interval debulking surgery. a Calculated by Chi-square test. b Calculated by Log-rank test using Kaplan-Meier method.
Table 3 Treatment details of primary recurrence in women receiving or not bevacizumab containing first line chemotherapy. Characteristics
All cases n (%)
Cases n (%)
Controls n (%)
Recurrence Complete secondary cytoreductive surgeryb No Yes Chemotherapy details in platinum sensitive recurrent patients Platinum-based Not Platinum-based Response to second line CHT (months, RECIST)c CR/PR SD/PD CR/PR to second line CHT (months, RECIST)c PFI b 6 6 ≤ PFI b 12 PFI ≥ 12 TTP for second line CHTd (Median and range, months) TTP for second line CHTd (Median and range, months) Primary debulking surgery +
155 (69.8)
39 (52.7)
116 (78.4)
81 (73.6) 29 (26.4)
27 (90.0) 3 (10.0)
54 (46.5) 26 (53.5)
0.016
78 (50.3) 77 (49.7)
19 (48.7) 20 (51.3)
59 (50.9) 57 (49.1)
0.481
56 (36.1) 99 (63.9)
9 (23.1) 30 (76.9)
47 (40.5) 69 (59.5)
0.055
4 (8.8) 18 (28.5) 34 (72.3)
1 (11.1) 3 (17.6) 5 (38.4)
3 (8.3) 15 (32.6) 29 (85.2)
0.995 0.350 0.002
adjuvant chemotherapy NACT followed by IDS Median TTP for second line CHTd (months) PFI b 6 months 6 months ≤ PFI
p valuea
6 (1–39)
5 (1–24)
8 (1–39)
0.041
8 (1–39)
6 (1–24)
9 (1–39)
0.175
4 (1−21)
4 (1–21)
6 (1−20) 0.023
3 (1–19) 9 (1–39)
3 (1–19) 5 (2–24)
3 (1–18) 10 (1–39)
0.922 0.006
PFI = platinum-free interval; TTP = time to progression (2nd line chemotherapy); CHT = chemotherapy; NACT = neoadjuvant chemotherapy; IDS = interval debulking surgery. a Calculated by Chi-square test. b Percentage calculated only considering women with recurrent disease and PFI ≥ 6. c Calculated according to RECIST criteria. CR = complete response, PR = partial response; SD = stable disease; PD = progressive disease. Percentages have been estimated dividing the number of patients showing CR/PR by the total number of women with recurrent disease in each specific PFI subgroup. d Calculated by Log-rank test using Kaplan-Meier method.
6
M. Petrillo et al. / Gynecologic Oncology xxx (2016) xxx–xxx
Fig. 3. Time to Progression to second line chemotherapy in Cases (solid line) and Controls (dotted line).
References [1] T.J. Perren, A.M. Swart, J. Pfisterer, et al., A phase 3 trial of bevacizumab in ovarian cancer, N. Engl. J. Med. 365 (2011) 2484–2496. [2] R.A. Burger, M.F. Brady, M.A. Bookman, et al., Incorporation of bevacizumab in the primary treatment of ovarian cancer, N. Engl. J. Med. 365 (2011) 2473–2483. [3] M. Markman, J. Markman, K. Webster, K. Zanotti, B. Kulp, G. Peterson, J. Belinson, Duration of response to second-line, platinum-based chemotherapy for ovarian cancer: implications for patient management and clinical trial design, J. Clin. Oncol. 22 (2004) 3120–3125. [4] J.A. Rauh-Hain, S.H. Guseh, K.M. Esselen, et al., Patterns of recurrence in patients treated with bevacizumab in the primary treatment of advanced epithelial ovarian cancer, Int. J. Gynecol. Cancer 23 (2013) 1219–1225. [5] M.J. Birrer, Y. Choi, M.F. Brady, et al., Retrospective analysis of candidate predictive tumor biomarkers (BMs) for efficacy in the GOG-0218 trial evaluating front-line carboplatin–paclitaxel (CP) ± bevacizumab (BEV) for epithelial ovarian cancer (EOC), J. Clin. Oncol. 33 (2015) (suppl; abstr 5505). [6] C. Gourley, A. McCavigan, T. Perren, et al., Molecular subgroup of high-grade serous ovarian cancer (HGSOC) as a predictor of outcome following bevacizumab, J. Clin. Oncol. 32 (5 s) (2014) (suppl; abstr 5502). [7] A. Fagotti, G. Vizzielli, F. Fanfani, et al., Introduction of staging laparoscopy in the management of advanced epithelial ovarian, tubal and peritoneal cancer: impact on prognosis in a single institution experience, Gynecol. Oncol. 131 (2013) 341–346. [8] A. Fagotti, G. Vizzielli, P. De Iaco, et al., A multicentric trial (Olympia-MITO 13) on the accuracy of laparoscopy to assess peritoneal spread in ovarian cancer, Am. J. Obstet. Gynecol. 209 (2013) 462.e1–462.e11. [9] M. Petrillo, A. Fagotti, G. Ferrandina, et al., Ovarian cancer patients with localized relapse: clinical outcome and prognostic factors, Gynecol. Oncol. 131 (2013) 36–41. [10] F. Fanfani, G. Monterossi, A. Fagotti, et al., Positron emission tomographylaparoscopy based method in the prediction of complete cytoreduction in platinum-sensitive recurrent ovarian cancer, Ann. Surg. Oncol. 22 (2015) 649–654. [11] A. Fagotti, F. Fanfani, C. Rossitto, et al., A treatment selection protocol for recurrent ovarian cancer patients: the role of FDG-PET/CT and staging laparoscopy, Oncology 75 (2008) 152–158.
[12] E.A. Eisenhauer, P. Therasse, J. Bogaerts, et al., New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1), Eur. J. Cancer 45 (2009) 228–247. [13] F.L. Kaplan, P. Meier, Non parametric estimation from incomplete observations, Am. J. Stat. Assoc. 53 (1958) 457–481. [14] N. Mantel, Evaluation of survival data and two new rank order statistics arising in its consideration, Cancer Chemother. Rep. 50 (1966) 163–170. [15] D.R. Cox, Regression models and life tables, J. R. Stat. Soc. 34 (1972) 197–220. [16] C. Aghajanian, B. Goff, L.R. Nycum, Y.V. Wang, A. Husain, S.V. Blank, Final overall survival and safety analysis of OCEANS, a phase 3 trial of chemotherapy with or without bevacizumab in patients with platinum-sensitive recurrent ovarian cancer, Gynecol. Oncol. 139 (2015) 10–16. [17] C. Aghajanian, S.V. Blank, B.A. Goff, et al., OCEANS: a randomized, double-blind, placebo-controlled phase III trial of chemotherapy with or without bevacizumab in patients with platinum-sensitive recurrent epithelial ovarian, primary peritoneal, or fallopian tube cancer, J. Clin. Oncol. 30 (2012) 2039–2045. [18] M.R. Stockler, F. Hilpert, M. Friedlander, et al., Patient-reported outcome results from the open-label phase III AURELIA trial evaluating bevacizumab-containing therapy for platinum-resistant ovarian cancer, J. Clin. Oncol. 32 (2014) 1309–1316. [19] E. Pujade-Lauraine, F. Hilpert, B. Weber, et al., Bevacizumab combined with chemotherapy for platinum-resistant recurrent ovarian cancer: the AURELIA open-label randomized phase III trial, J. Clin. Oncol. 32 (2014) 1302–1308. [20] P.J. Hoskins, M. Fung-Kee-Fung, D. Miller, W. Gotlieb, Time for a level playing field: inequalities in regulatory/approval processes—the example of bevacizumab in epithelial ovarian cancer, J. Clin. Oncol. 33 (2015) 1539–1542. [21] M. Petrillo, I. Paris, G. Vizzielli, et al., Neoadjuvant chemotherapy followed by maintenance therapy with or without bevacizumab in unresectable high-grade serous ovarian cancer: a case-control study, Ann. Surg. Oncol. 22 (Suppl. 3) (2015) 952–958. [22] H. Dalton, Role of macrophages in adaptive resistance to anti-vegf therapy, 2014 (UT GSBS Dissertations and Theses (Open Access). Paper 472). [23] M. Jinushi, Y. Komohara, Tumor-associated macrophages as an emerging target against tumors: creating a new path from bench to bedside, Biochim. Biophys. Acta 2015 (1855) 123–130. [24] T. Mitamura, C. Gourley, A.K. Sood, Prediction of anti-angiogenesis escape, Gynecol. Oncol. 141 (2016) 80–85.
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Society Position Statements/White Papers
Impact of bevacizumab containing first line chemotherapy on recurrent disease in epithelial ovarian cancer: A case-control study M. Petrillo a, G. Amadio a, V. Salutari a, I. Paris a, M.G. Di Stefano a, G. Ferandina b,c, G. Scambia a, A. Fagotti d,e,⁎ a
Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Catholic University of the Sacred Heart, Rome, Italy Department of Medicine and Health Sciences, University of Molise, Campobasso, Italy Gynecologic Oncology Unit, Foundation “Policlinico Universitario A. Gemelli”, Rome, Italy d Department of Obstetrics and Gynecology, Catholic University of the Sacred Heart, Rome, Italy e University of Perugia, Italy b c
H I G H L I G H T S • Incorporation of bevacizumab into upfront regimens prolongs PFI in AOC patients. • However, it is associated with wider presentation of relapse, and lower rate of complete SCS. • TTP to second line chemotherapy was shorter in women with platinum-sensitive relapse initially treated with bevacizumab.
a r t i c l e
i n f o
Article history: Received 2 April 2016 Received in revised form 7 May 2016 Accepted 16 May 2016 Available online xxxx Keywords: Bevacizumab Recurrent ovarian cancer Secondary cytoreductive surgery Response to second-line chemotherapy
a b s t r a c t Objective. To evaluate the timing and pattern of relapse, and duration of response to second line chemotherapy in advanced ovarian cancer (AOC) patients treated with first line carboplatin-paclitaxel chemotherapy with or without bevacizumab. Patients and methods. This is a case-control study including 222 AOC patients. Seventy-four women treated with first line carboplatin-paclitaxel-bevacizumab chemotherapy (Cases) were matched based on laparoscopic predictive index value, and residual tumor at first surgery with 148 AOC patients treated with carboplatinpaclitaxel. Distribution of pattern of relapse, and response to second line chemotherapy was compared between the two groups. Time to Progression (TTP) for second line chemotherapy was also analyzed for study purpose. Results. Median platinum-free interval (PFI) was 16 months (range 2–65) in Cases, compared with 9 months (1–83) in Controls (p-value = 0.001). Twenty patients (51.3%) among Cases showed recurrence in multiple anatomic sites, compared with 31 (31.9%) in the Control group (p-value = 0.035). Peritoneal recurrence occurred as diffuse in 30 Cases (96.8%), and 60 Controls (82.2%; p-value = 0.046). Secondary cytoreductive surgery (SCS) was successfully completed in 53.5% of Controls compared to 10.0% of Cases (p-value = 0.016). In women with fully platinum-sensitive relapse, response rate to second line chemotherapy was 85.2% in Controls, compared to 38.4% in Cases (p-value = 0.002). Finally, Cases showed a shorter TTP, compared to Controls (5 months vs 8 months; p-value = 0.041). Conclusions. Incorporation of bevacizumab into upfront regimens prolongs PFI in AOC patients, but is associated with wider presentation of relapse, lower rate of complete SCS, and shorter TTP to second line chemotherapy in women with platinum-sensitive disease. © 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction In the past decade, the results of two pivotal randomized phase III clinical trials demonstrated a significant improvement of progression⁎ Corresponding author at: Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Catholic University of the Sacred Heart, Largo A. Gemelli, 8, 00168 Rome, Italy. E-mail address: [email protected] (A. Fagotti).
free survival (PFS), in patients with advanced ovarian cancer (AOC) receiving bevacizumab as part of first line treatment [1,2]. Based on these data, in December 2011, Avastin received European regulatory approval for use in combination with carboplatin-paclitaxel as upfront chemotherapy regimen in OC patients with advanced disease. The observed prolongation of PFS should be considered a relevant therapeutic achievement, since it delays the onset of physical symptoms associated with progressive disease, as well as the need to start a second-line chemotherapy. Furthermore, the prolongation of PFS may
http://dx.doi.org/10.1016/j.ygyno.2016.05.017 0090-8258/© 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
2
M. Petrillo et al. / Gynecologic Oncology xxx (2016) xxx–xxx
in principle increase the chance of response to further platinum-based treatments [3], thus offering a potential long-term benefit. However, in this encouraging scenario, it is reasonable to hypothesize that the incorporation of bevacizumab into first-line treatment may exert an additional selective pressure on OC cells, with a potentially relevant impact on the natural history of this malignancy. This hypothesis is also supported by recently published retrospective data, which demonstrated a higher rate of lung and pleural recurrence in AOC patients receiving bevacizumab as part of primary treatment [4], thus suggesting that more efforts should be done to definitely clarify how bevacizumab will change the clinical features of OC. Furthermore, despite initial promising results [5,6], it remains to be validated an effective signature able to properly identify at diagnosis those AOC patients who may real benefit from first line bevacizumab-based chemotherapy. For these reasons, we focused our attention on the impact of upfront bevacizumab containing regimens on recurrent disease, analyzing the timing and pattern of relapse, as well as the rate and duration of response to second line chemotherapy in a single-Institutional case control study including AOC patients treated with first line carboplatinpaclitaxel chemotherapy with or without bevacizumab. 2. Patients and methods 2.1. Study groups The investigational arm (Cases) included 74 patients with AOC admitted at the Gynaecologic Oncology Unit of the Catholic University of Rome and Campobasso between March 2010 and February 2013 treated, after joint evaluation with the oncologists' team, with the following regimen: carboplatin 5 AUC/paclitaxel 175 mg per square meter of body-surface area/bevacizumab 15 mg/Kg every three weeks, followed by maintenance therapy with bevacizumab 15 mg/Kg until disease progression or for a maximum of 15 months. The Control arm included a consecutive series of 148 AOC patients, admitted at our Institution between January 2010 and December 2012, treated with carboplatin 5 AUC/paclitaxel 175 mg per square meter of body-surface area every three weeks. In both groups, patients were selected for primary debulking surgery (PDS), or neoadjuvant chemotherapy based on our Institutional laparoscopy-based algorithm [7], and to avoid imbalance between the two groups in term of initial disease extension and clinical features the Cases were matched with the Controls according to the laparoscopic predictive index value [7,8], and residual tumor at first surgery, using Stata software version 11.0 (Stata Corp, College station, TX). Furthermore, to maximize the power of the study, Cases and Controls were matched in a 1:2 ratio. All women gave a written informed consent for their data to be collected and analyzed for scientific purpose. The Institutional Review Board approved the study.
All patients showing platinum-sensitive recurrent disease were submitted to FDG-PET/CT scan, and staging-laparoscopy, to assess the chance of optimal secondary cytoreduction (SCS) [10,11]. All patients judged as suitable for optimal SCS received an attempt of surgical debulking. Patients showing platinum-sensitive recurrent disease received second-line platinum-based chemotherapy; on the other hand, women showing platinum-resistant relapse were treated with nonplatinum agents including weekly paclitaxel, pegylated liposomal doxorubicin, and gemcitabine. The group of cases developing recurrence with a PFI between 6, and 12 months were triaged to receive platinum versus non-platinum regimens, based on a case-specific decision making process. Clinical response to second line chemotherapy was assigned according to RECIST criteria [12], and the response has been classified as complete (CR)/partial (PR), or stable (SD)/progressive (PD). In order to fully evaluate the efficacy of second line treatment, we analyzed in our study also the Time to Progression (TTP), which was defined as the time from beginning of second line chemotherapy to further progressive disease. Data analysis regarding overall survival was not attempted due to the duration of follow-up period. 2.3. Statistical analysis Differences between Cases and Controls in term of clinicopathological features at diagnosis, and at the time of relapse were analyzed using the Pearson Chi-square exact test and Kruskall-Wallis test, as appropriate. Medians and life tables were computed using the product limit estimate by Kaplan–Meier method [13], and the log-rank test was used to assess the statistical significance [14]. Cox's regression model with stepwise variable selection [15] was used to analyse the role of clinical-pathological parameters, and treatment details as predictors of TTP for second line chemotherapy. All statistical calculations were performed using the Stata software version 11.0 (Stata Corp, College station, TX). 3. Results 3.1. Clinico-pathological features at diagnosis As a consequence of matching process, no differences were observed between Cases and Controls in term of disease extension at diagnosis evaluated as laparoscopic predictive index value, and residual tumor at first surgery (Table 1). Around 90% of AOC patients showed highgrade serous histology, and FIGO Stage IIIC at initial diagnosis, without differences between the two groups (Table 1). Median CA125 levels were 792 IU (range 10–7654) in the overall population, with similar levels in Cases and Controls (p-value = 0.607). Median age at diagnosis of women treated with first line bevacizumab-containing chemotherapy was 54 years (range 26–76), which appears slightly lower compared with Controls (median 59, range 29–86; p-value = 0.069).
2.2. Clinical data
3.2. Timing of recurrent disease
For all patients included in the final analysis, data regarding timing, pattern, and treatment details at the time of relapse were prospectively recorded, and retrospectively analyzed for study purpose. In particular, the timing of recurrent disease was evaluated using the PFI defined as the time elapsed between the date of completion of platinum-based first line chemotherapy and recurrence. The pattern of relapse was classified as peritoneal, parenchymal, and lymph-nodal based on the site of disease [9]. Furthermore, we defined the type of recurrence as single in presence of only one anatomic site involved, or multiple when relapse diffusion occurred in more than one site. Finally, the extension of intraperitoneal recurrence was classified as localized in presence of no more than three nodules, or diffused when a wider disease spread was observed [9].
After a median follow-up of 32 months (Cases 25 months, Controls 32 months; p-value = 0.785), we observed recurrent disease in 39 Cases (52.7%), and 116 Controls (78.4%; p-value = 0.001; Fig. 1). Median PFI was 16 months (range 2–65) in Cases, compared with 9 months (4–83) in the Control group (p-value = 0.001; Fig. 2). AOC patients initially treated with PDS showed a median PFI of 10 months (1–83) in Controls, compared with 19 months (2–65) in Cases (p-value = 0.001, Table 2). Similarly, in the group of women treated with upfront NACT followed by interval debulking surgery, we observed a longer PFI in patients treated with bevacizumab (10 months) compared with Controls (7 months; p-value = 0.032; Table 2). The distribution of relapse according with PFI in the two groups has been reported in Table 2. In particular, we observed platinum-resistant relapse in 45
M. Petrillo et al. / Gynecologic Oncology xxx (2016) xxx–xxx
3
Controls n (%)
p valuea
74 (33.3) 54 (26–76)
148 (66.7) 59 (29–86)
0.069
202 (90.9) 11 (4.9) 9 (4.2)
64 (86.4) 5 (6.8) 5 (6.8)
138 (93.2) 6 (4.1) 4 (2.7)
0.134
only in the subgroup of women with platinum-sensitive recurrent disease. Furthermore, as presented in Table 3, TTP was analyzed according to initial treatment strategy (NACT or PDS), and in the group of women who received PDS, duration of TTP was 6 months (1–24) in Cases, and 9 months (1–39) in Controls (p-value = 0.175). Similarly, also in patients initially selected for NACT, we observed a longer median TTP in Controls (6 months) compared with Cases (4 months, p-value = 0.023; Table 3). Finally, at multivariate analysis, the administration of bevacizumab in the context of first line treatment confirmed its independent negative prognostic role for duration of TTP (p value = 0.047; Supplementary Table 1).
85 (38.3) 51 (20.0) 86 (38.7)
28 (37.8) 17 (23.0) 29 (39.2)
57 (38.5) 34 (23.0) 57 (38.5)
0.994
4. Discussion
203 (91.4) 19 (8.6)
67 (90.5) 7 (9.5)
136 (91.9) 12 (8.1)
0.800
Table 1 Distribution of patients' clinico-pathological characteristics at diagnosis in cases and controls. Characteristics
All pts
All Age (median and range, years)b Tumor histotypes High-grade serous Endometrioid Clear cell LPS-PIV at diagnosis b4 4–6 ≥8 FIGO stage IIIC IV Residual tumor at first surgery Complete ≤1 cm NACT CA125 (median and range, UI/ml)b
222 58 (26–86)
Cases n (%)
132 (59.5) 44 (59.5) 88 (59.5) 20 (9.0) 6 (8.1) 14 (9.5) 70 (31.5) 24 (32.4) 46 (31.1) 0.938 792 (10–7654) 707 (10–9842) 852 (24–7654) 0.667
LPS-PIV = laparoscopic predictive index value; NACT = neoadjuvant chemotherapy. a Calculated by Chi-square test. b Calculated by Kruskal-Wallis non-parametric test.
patients (29.0%), partially platinum-sensitive recurrent in 63 women (40.6%), and fully platinum-sensitive disease in 47 (30.4%) cases, without differences between the two groups (p-value = 0.636; Table 2). 3.3. Pattern of recurrent disease Focusing on pattern of disease presentation, 20 patients (51.3%) among Cases showed involvement of multiple anatomic sites, compared with 31 (31.9%) women in the Control group (p-value = 0.035). This wider recurrent spread was sustained by higher incidence of peritoneal and lymph nodal relapse in Cases compared with Controls (Table 2). Finally, peritoneal recurrence occurred more frequently as diffuse disease in patients treated with bevacizumab-based first line regimen (30 patients, 96.8%), compared with women receiving conventional treatment (60 patients, 82.2%; p-value = 0.046). 3.4. SCS and response to second line chemotherapy As a consequence of differences in term of disease spread, SCS was successfully completed in a larger group of Controls (53.5%) compared to Cases (10.0%, p-value = 0.016; Table 3). Overall, platinum-based second line chemotherapy was administered in 78 patients (50.3%), while 77 (49.7%) women received non-platinum containing regimens, without differences between the two groups (p-value = 0.481). Response to second line chemotherapy (CR + PR) was observed in 9 Cases (23.1%), and 47 Controls (40.5%), with a trend toward a statistically significant difference (p-value = 0.055; Table 3). Furthermore, when response rate was analyzed according with duration of PFI, we documented a statistically significant difference in favour of Controls (85.2%) compared to Cases (38.4%, p-value = 0.002; Table 3) in the subgroup of women with fully platinum-sensitive relapse. 3.5. Time to progression for second line chemotherapy At Kaplan-Meier analysis, Cases showed a shorter median TTP, compared to Controls (5 months vs 8 months; p-value = 0.041; Fig. 3). However, when duration of TTP was analyzed according with PFI, median TTP resulted significantly longer in Controls compared to Cases (5 months vs 10 months; p-value = 0.006; Supplementary Fig. 1),
The observation of a longer PFS in women with AOC receiving bevacizumab-containing first line chemotherapy certainly represents a relevant achievement in the management of epithelial OC; furthermore, a survival benefit related to bevacizumab administration was demonstrated also in other clinical settings including women with platinumsensitive [16,17], and resistant recurrent disease [18,19]. However, even if the overall efficacy of bevacizumab appears unquestionable, several issues are still open, including: the best clinical setting in which the drug should be administered to maximize overall survival [20], the most appropriate duration of maintenance therapy (AGO-OVAR17; NCT01462890; ROSiA NCT01239732), and whether bevacizumab administration could give a benefit beyond progression (MITO-16/ MaNGO OV-2, NCT01802749). In this continuously changing horizon, we present here, the first single Institution case-control study comparing the features of recurrent disease in a large series of AOC patients treated with or without bevacizumab as first line treatment. The accurate matching in term of disease extension, using as measure the laparoscopic index value, and the exact balancing of cases and controls in term of residual tumor at first surgery represents major strengthens of our study, thus ultimately supporting the reliability of our findings. For instance, we confirmed in our series that the incorporation of bevacizumab into first line treatment ensures an around seven months prolongation of PFI, in the overall population of AOC patients [1–2], and we confirmed previously published data suggesting a bevacizumabrelated benefit in term of PFS also in patients treated with NACT [21]. However, despite the longer PFS in the group of women treated with bevacizumab, we did not observe differences in the distribution of patients with fully-sensitive, partially-sensitive or platinum-resistant relapse between the Cases and Controls. Based on these data, it may be hypothesized that the major contribution of bevacizumab upfront administration is mainly represented by an overall reduction of recurrence rate, rather than a prolongation of PFI. On the other hand, as previously reported in a retrospective analysis [21], we documented differences in term of pattern of disease presentation between the two groups. In particular, we observed a more frequent involvement of multiple anatomic sites in relapsed AOC patients initially treated with bevacizumab-based regimens. Furthermore, focusing on intraperitoneal recurrence we documented a statistically significant higher incidence of diffuse disease in Cases compared with Controls. Far from drawing definitive conclusions regarding the survival impact of the more unfavourable pattern of relapse in AOC patients initially treated with bevacizumab, our data may contribute to explain the lack of a clear benefit in term of overall survival emerged in Phase III randomized clinical trials [1–2]. Indeed, the observed disease diffusion at the time of recurrence carries on relevant therapeutic implications. In particular, as a consequence of the more frequent involvement of multiple anatomic sites, and of the wider diffusion of peritoneal disease, SCS was successfully attempted in only 10% of women previously treated with bevacizumab, compared with around 50% of women receiving optimal SCS in the Control group.
4
M. Petrillo et al. / Gynecologic Oncology xxx (2016) xxx–xxx
Fig. 1. Flow-chart of our study population.
Another relevant result of our analysis is represented by trend toward a lower response rate, and a three months shorter statistically significant TTP to second line chemotherapy in favour of Controls compared with Cases. The differences in term of response rate appear of higher magnitude, reaching statistical significance, in patients with fully-platinum sensitive recurrent disease. Similarly, the longer TTP to second line chemotherapy in favour of Controls documented in the overall population was not observed in the group of platinum-resistant disease, being however confirmed in patients with a PFI ≥ 6 months. It could be argued that the above mentioned differences between Cases and Controls in term of response rate, ant TTP to second line chemotherapy may be the consequence of wider pattern of disease presentation, and a lower percentage of SCS in the group of Cases. However, taken together our findings, and previously published experiences [4], suggest that the incorporation of bevacizumab into first line regimens, even ensuring an overall PFS benefit, influences the disease natural history, and in particular the clinical features, and management of recurrent disease.
In this context, recently published data demonstrated in preclinical models a potential role of tumor-associated macrophages in driving resistance to antiangiogenic agents [22–23], thus emphasizing that further biological treatment strategies should be developed to reduce the clinical impact of the mechanisms involved in determining the acquisition of adaptive resistance to antiangiogenic agents [24]. In conclusions, despite the exact matching in term of disease extension, and residual tumor at first surgery, the retrospective nature, and the small sample size certainly represent relevant study limitations. However, awaiting results from randomized clinical trials, our data suggest for the first time, in the context of a case-control study, that the incorporation of bevacizumab into upfront regimens prolongs PFI in AOC patients, but is associated with a more aggressive behaviour of recurrent disease, including a lower percentage of patients suitable for SCS, and a shorter TTP to second line chemotherapy in women with platinum-sensitive disease. Larger studies are needed to confirm these data, and a multi-Institutional commitment is urgently required to identify molecular signatures able to predict response to bevacizumab first line treatment, and to clarify the biological effects of chronic antiangiogenic exposure of OC cells.
M. Petrillo et al. / Gynecologic Oncology xxx (2016) xxx–xxx
5
Fig. 2. Platinum-free interval in Cases (solid line) and Controls (dotted line).
Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ygyno.2016.05.017.
Conflict of interest statement The authors declare that there are no conflicts of interest.
Table 2 Timing and pattern of recurrent disease in women receiving or not bevacizumab containing first line chemotherapy. Characteristics
All cases n (%)
Recurrence PFIb (median and range, months) PFIb (median and range, months) Primary debulking surgery +
155 (69.8) 39 (52.7) 116 (78.4) 0.001 11 (1–83) 16 (2–65) 9 (1–83) 0.001
adjuvant chemotherapy NACT followed by IDS Timing of relapse (months) PFI b 6 months 6 ≤ PFI b 12 months PFI ≥ 12 months Site of recurrence Parenchyma (lung, spleen, liver) Lymph nodes Peritoneum Type of recurrence Single site Multiple sites Extension of intraperitoneal recurrence Localized Diffused
Cases n (%)
Controls n (%)
p valuea
14 (1–83) 19 (2–65)
10 (1–83) 0.001
8 (1–36) 10 (1–36)
7 (1–18) 0.032
45 (29.0) 63 (40.6) 47 (30.4)
9 (23.1) 17 (43.6) 13 (33.3)
36 (31.0) 46 (39.7) 34 (29.3)
18 (11.4) 50 (32.1) 104 (66.5)
6 (15.4) 20 (51.3) 31 (79.5)
12 (10.3) 30 (25.9) 73 (62.9)
98 (63.2) 57 (36.8)
19 (48.7) 20 (51.3)
79 (68.1) 37 (31.9)
17 (10.9) 138 (89.1)
1 (3.2) 30 (96.8)
13 (17.8) 60 (82.2)
0.636 0.279 0.004 0.041
0.035
0.046
PFI = platinum-free interval; NACT = neoadjuvant chemotherapy; IDS = interval debulking surgery. a Calculated by Chi-square test. b Calculated by Log-rank test using Kaplan-Meier method.
Table 3 Treatment details of primary recurrence in women receiving or not bevacizumab containing first line chemotherapy. Characteristics
All cases n (%)
Cases n (%)
Controls n (%)
Recurrence Complete secondary cytoreductive surgeryb No Yes Chemotherapy details in platinum sensitive recurrent patients Platinum-based Not Platinum-based Response to second line CHT (months, RECIST)c CR/PR SD/PD CR/PR to second line CHT (months, RECIST)c PFI b 6 6 ≤ PFI b 12 PFI ≥ 12 TTP for second line CHTd (Median and range, months) TTP for second line CHTd (Median and range, months) Primary debulking surgery +
155 (69.8)
39 (52.7)
116 (78.4)
81 (73.6) 29 (26.4)
27 (90.0) 3 (10.0)
54 (46.5) 26 (53.5)
0.016
78 (50.3) 77 (49.7)
19 (48.7) 20 (51.3)
59 (50.9) 57 (49.1)
0.481
56 (36.1) 99 (63.9)
9 (23.1) 30 (76.9)
47 (40.5) 69 (59.5)
0.055
4 (8.8) 18 (28.5) 34 (72.3)
1 (11.1) 3 (17.6) 5 (38.4)
3 (8.3) 15 (32.6) 29 (85.2)
0.995 0.350 0.002
adjuvant chemotherapy NACT followed by IDS Median TTP for second line CHTd (months) PFI b 6 months 6 months ≤ PFI
p valuea
6 (1–39)
5 (1–24)
8 (1–39)
0.041
8 (1–39)
6 (1–24)
9 (1–39)
0.175
4 (1−21)
4 (1–21)
6 (1−20) 0.023
3 (1–19) 9 (1–39)
3 (1–19) 5 (2–24)
3 (1–18) 10 (1–39)
0.922 0.006
PFI = platinum-free interval; TTP = time to progression (2nd line chemotherapy); CHT = chemotherapy; NACT = neoadjuvant chemotherapy; IDS = interval debulking surgery. a Calculated by Chi-square test. b Percentage calculated only considering women with recurrent disease and PFI ≥ 6. c Calculated according to RECIST criteria. CR = complete response, PR = partial response; SD = stable disease; PD = progressive disease. Percentages have been estimated dividing the number of patients showing CR/PR by the total number of women with recurrent disease in each specific PFI subgroup. d Calculated by Log-rank test using Kaplan-Meier method.
6
M. Petrillo et al. / Gynecologic Oncology xxx (2016) xxx–xxx
Fig. 3. Time to Progression to second line chemotherapy in Cases (solid line) and Controls (dotted line).
References [1] T.J. Perren, A.M. Swart, J. Pfisterer, et al., A phase 3 trial of bevacizumab in ovarian cancer, N. Engl. J. Med. 365 (2011) 2484–2496. [2] R.A. Burger, M.F. Brady, M.A. Bookman, et al., Incorporation of bevacizumab in the primary treatment of ovarian cancer, N. Engl. J. Med. 365 (2011) 2473–2483. [3] M. Markman, J. Markman, K. Webster, K. Zanotti, B. Kulp, G. Peterson, J. Belinson, Duration of response to second-line, platinum-based chemotherapy for ovarian cancer: implications for patient management and clinical trial design, J. Clin. Oncol. 22 (2004) 3120–3125. [4] J.A. Rauh-Hain, S.H. Guseh, K.M. Esselen, et al., Patterns of recurrence in patients treated with bevacizumab in the primary treatment of advanced epithelial ovarian cancer, Int. J. Gynecol. Cancer 23 (2013) 1219–1225. [5] M.J. Birrer, Y. Choi, M.F. Brady, et al., Retrospective analysis of candidate predictive tumor biomarkers (BMs) for efficacy in the GOG-0218 trial evaluating front-line carboplatin–paclitaxel (CP) ± bevacizumab (BEV) for epithelial ovarian cancer (EOC), J. Clin. Oncol. 33 (2015) (suppl; abstr 5505). [6] C. Gourley, A. McCavigan, T. Perren, et al., Molecular subgroup of high-grade serous ovarian cancer (HGSOC) as a predictor of outcome following bevacizumab, J. Clin. Oncol. 32 (5 s) (2014) (suppl; abstr 5502). [7] A. Fagotti, G. Vizzielli, F. Fanfani, et al., Introduction of staging laparoscopy in the management of advanced epithelial ovarian, tubal and peritoneal cancer: impact on prognosis in a single institution experience, Gynecol. Oncol. 131 (2013) 341–346. [8] A. Fagotti, G. Vizzielli, P. De Iaco, et al., A multicentric trial (Olympia-MITO 13) on the accuracy of laparoscopy to assess peritoneal spread in ovarian cancer, Am. J. Obstet. Gynecol. 209 (2013) 462.e1–462.e11. [9] M. Petrillo, A. Fagotti, G. Ferrandina, et al., Ovarian cancer patients with localized relapse: clinical outcome and prognostic factors, Gynecol. Oncol. 131 (2013) 36–41. [10] F. Fanfani, G. Monterossi, A. Fagotti, et al., Positron emission tomographylaparoscopy based method in the prediction of complete cytoreduction in platinum-sensitive recurrent ovarian cancer, Ann. Surg. Oncol. 22 (2015) 649–654. [11] A. Fagotti, F. Fanfani, C. Rossitto, et al., A treatment selection protocol for recurrent ovarian cancer patients: the role of FDG-PET/CT and staging laparoscopy, Oncology 75 (2008) 152–158.
[12] E.A. Eisenhauer, P. Therasse, J. Bogaerts, et al., New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1), Eur. J. Cancer 45 (2009) 228–247. [13] F.L. Kaplan, P. Meier, Non parametric estimation from incomplete observations, Am. J. Stat. Assoc. 53 (1958) 457–481. [14] N. Mantel, Evaluation of survival data and two new rank order statistics arising in its consideration, Cancer Chemother. Rep. 50 (1966) 163–170. [15] D.R. Cox, Regression models and life tables, J. R. Stat. Soc. 34 (1972) 197–220. [16] C. Aghajanian, B. Goff, L.R. Nycum, Y.V. Wang, A. Husain, S.V. Blank, Final overall survival and safety analysis of OCEANS, a phase 3 trial of chemotherapy with or without bevacizumab in patients with platinum-sensitive recurrent ovarian cancer, Gynecol. Oncol. 139 (2015) 10–16. [17] C. Aghajanian, S.V. Blank, B.A. Goff, et al., OCEANS: a randomized, double-blind, placebo-controlled phase III trial of chemotherapy with or without bevacizumab in patients with platinum-sensitive recurrent epithelial ovarian, primary peritoneal, or fallopian tube cancer, J. Clin. Oncol. 30 (2012) 2039–2045. [18] M.R. Stockler, F. Hilpert, M. Friedlander, et al., Patient-reported outcome results from the open-label phase III AURELIA trial evaluating bevacizumab-containing therapy for platinum-resistant ovarian cancer, J. Clin. Oncol. 32 (2014) 1309–1316. [19] E. Pujade-Lauraine, F. Hilpert, B. Weber, et al., Bevacizumab combined with chemotherapy for platinum-resistant recurrent ovarian cancer: the AURELIA open-label randomized phase III trial, J. Clin. Oncol. 32 (2014) 1302–1308. [20] P.J. Hoskins, M. Fung-Kee-Fung, D. Miller, W. Gotlieb, Time for a level playing field: inequalities in regulatory/approval processes—the example of bevacizumab in epithelial ovarian cancer, J. Clin. Oncol. 33 (2015) 1539–1542. [21] M. Petrillo, I. Paris, G. Vizzielli, et al., Neoadjuvant chemotherapy followed by maintenance therapy with or without bevacizumab in unresectable high-grade serous ovarian cancer: a case-control study, Ann. Surg. Oncol. 22 (Suppl. 3) (2015) 952–958. [22] H. Dalton, Role of macrophages in adaptive resistance to anti-vegf therapy, 2014 (UT GSBS Dissertations and Theses (Open Access). Paper 472). [23] M. Jinushi, Y. Komohara, Tumor-associated macrophages as an emerging target against tumors: creating a new path from bench to bedside, Biochim. Biophys. Acta 2015 (1855) 123–130. [24] T. Mitamura, C. Gourley, A.K. Sood, Prediction of anti-angiogenesis escape, Gynecol. Oncol. 141 (2016) 80–85.