- Email: [email protected]
Prognostic value of 18F-fluorodeoxyglucose uptake in pelvic lymph nodes in patients with cervical cancer treated with definitive chemoradiotherapy
Gynecologic Oncology 137 (2015) 40–46
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Prognostic value of 18F-fluorodeoxyglucose uptake in pelvic lymph nodes in patients with cervical cancer treated with definitive chemoradiotherapy Cem Onal a,⁎, Ozan C. Guler a, Mehmet Reyhan b, Ali Fuat Yapar b a b
Department of Radiation Oncology, Baskent University Faculty of Medicine, Adana, Turkey Department of Nuclear Medicine, Baskent University Faculty of Medicine, Ankara, Turkey
H I G H L I G H T S • The prognostic value of FDG uptake in regional lymph nodes for cervical cancer patients has been rarely studied. • Pelvic lymph node SUVmax, post-therapy metabolic response and para-aortic lymph node metastases were significant prognostic factors for survival. • High-risk patients may benefits from additional treatments such as systemic chemotherapy.
a r t i c l e
i n f o
Article history: Received 24 October 2014 Accepted 23 January 2015 Available online 29 January 2015 Keywords: Cervical cancer Chemoradiotherapy Lymph node metastasis Positron emission tomography
a b s t r a c t Purpose. To evaluate the prognostic significance of the maximum standardized uptake (SUVmax) value for pelvic lymph nodes in patients with cervical cancer and its impact on treatment response, disease control, and survival. Methods. Ninety-three patients with pelvic or para-aortic metastasis detected by PET/CT and treated with definitive chemoradiotherapy were evaluated. The impact of pelvic lymph node SUVmax on prognostic factors and treatment outcomes was assessed. Results. The size and SUVmax of pelvic lymph nodes were significantly correlated (r = 0.859; p b 0.001). Patients with pelvic and para-aortic lymph node metastases had significantly higher SUVmax values for both primary tumor (23.4 ± 9.2 vs. 18.5 ± 7.3; p = 0.01) and pelvic lymph nodes (11.4 ± 4.6 vs. 7.4 ± 3.8; p = 0.001). Patients with pelvic lymph node SUVmax ≥ 7.5 had significantly higher primary tumor SUVmax, larger pelvic lymph nodes, higher rates of para-aortic lymph node metastasis, and lower post-therapy complete response rates. Overall survival (OS) and disease-free survival (DFS) rates were significantly higher in patients with SUVmax b 7.5 compared to patients with SUVmax ≥ 7.5. In a multivariate analysis, pelvic lymph node SUVmax and post-therapy metabolic response were significant prognostic factors for both OS and DFS for all patients, but no significant prognostic factors were found in pelvic lymph node metastasis only. Conclusions. Patients with highly FDG-avid pelvic lymph nodes have a higher risk of disease recurrence with worse survival. Identification of these patients may assist in the evaluation of the clinical benefits of additional treatments. © 2015 Elsevier Inc. All rights reserved.
Introduction Cervical cancer is the second-most frequently diagnosed gynecological cancer worldwide. Almost half of such patients present with advanced disease at the time of initial diagnosis. Some prognostic factors have been associated with clinical outcome, including young ⁎ Corresponding author at: Baskent University Faculty of Medicine, Adana Research and Treatment Centre, Department of Radiation Oncology, 01120 Adana, Turkey. Fax: +90 322 3444445. E-mail address: [email protected] (C. Onal).
http://dx.doi.org/10.1016/j.ygyno.2015.01.542 0090-8258/© 2015 Elsevier Inc. All rights reserved.
age, pretreatment stage, and lymph node status [1,2]. Pelvic and/or para-aortic lymph node metastases have a strong impact on the prognosis and management of treatment in patients with cervical cancer. The lymphatic spread of cervical cancer occurs mainly in an orderly fashion, from pelvic to para-aortic lymph nodes to supraclavicular and mediastinal lymph nodes in advanced stages. Additionally, pelvic lymph node metastasis increases the risk of para-aortic lymph node metastasis [3]. 18 F-fluorodeoxyglucose (FDG) positron emission tomography (PET) integrated with computed tomography (PET/CT) incorporates metabolic tumor function with anatomic localization. PET/CT has become an increasingly important component of tumor staging by virtue of its
C. Onal et al. / Gynecologic Oncology 137 (2015) 40–46
ability to identify lymph node involvement, distant disease, and recurrent disease in patients with cervical cancer [4]. Some studies have also assessed the relationship between FDG uptake and outcome in cervical cancer [5,6]. Previously, we found that the maximum standardized uptake value (SUVmax) for the primary tumor was correlated with increased tumor size and lymph node metastasis rates, persistence of tumor after treatment, and worse overall survival—even after a complete metabolic response [7,8]. However, the prognostic value of FDG uptake in regional lymph nodes for cervical cancer patients has been rarely studied, and the predictive value of FDG uptake in pelvic lymph nodes remains unclear [9,10]. In a study involving patients with cervical cancer, Yen et al. [10] found that an SUVmax N 3.3 in para-aortic lymph nodes was predictive of worse overall survival in 14 of 70 patients with para-aortic lymph node metastasis, but found no impact of the pelvic lymph node SUVmax on overall survival (OS). In another study, Kidd et al. [9] demonstrated that the pelvic lymph node SUVmax was a prognostic biomarker and predictor of treatment outcome in 83 patients with cervical cancer. Given these recent studies suggesting that the SUVmax of involved lymph nodes is predictive of disease control and OS, and our previous research demonstrating the prognostic value of the primary tumor SUVmax as determined by FDG-PET [8], we sought to evaluate the prognostic significance of the pelvic lymph node SUVmax in patients with cervical cancer. Additionally, we analyzed the impact of the pelvic lymph node SUVmax on treatment response, disease control, and survival. Materials and methods Patients The study population consisted of 183 patients with biopsy-proven cervical cancer treated with curative intent between November 2006 and August 2013; 93 patients (51%) had pelvic or para-aortic metastasis detected by PET/CT. Approval was obtained from the institutional review board for this retrospective outcome analysis. All patients underwent routine clinical staging, including recording of medical history, physical and gynecological examinations, blood chemistry tests, complete blood count, and magnetic resonance imaging (MRI) or CT of the abdomen and pelvis. The patients were staged according to the International Federation of Gynecology and Obstetrics (FIGO) staging system. All patients had also undergone PET/CT for initial diagnosis, staging, and radiotherapy (RT) planning. PET/CT technique The patients were imaged using a dedicated PET/CT system (Discovery-STE 8; General Electric Medical Systems, Milwaukee, WI) as previously described [11]. The patients fasted for at least 6 h before intravenous administration of 370 to 555 MBq (10–15 mCi) FDG. Preinjection blood glucose levels were measured to ensure that they were below 150 mg/dL. During the distribution phase, the patients laid supine in a quiet room. Combined image acquisition began 60 min after FDG injection. The patients were scanned on a flat-panel, carbon-fiber composite table insert. First, an unenhanced CT scan (5-mm slice thickness) from the base of the skull to the inferior border of the pelvis was acquired using a standardized protocol (140 kV and 80 mA). The subsequent PET scan was acquired in three-dimensional (3D) mode from the base of the skull to the inferior border of the pelvis (6 to 7 bed positions, 3 min per position) without repositioning the patient on the table. CT and PET images were acquired with the patient breathing shallowly. Attenuation was corrected using the CT images. Areas of FDG uptake were categorized as malignant based on location, intensity, shape, size, and visual correlation with CT images to differentiate physiologic uptake from pathologic uptake.
41
A lymph node was considered PET-positive if its FDG uptake was greater than blood pool activity or surrounding background tissues, depending on the size of the node. The maximum long axis of a pelvic lymph node with FDG uptake was recorded at co-registered PET/CT images. Patients were eligible for this study if at least one FDG-avid pelvic lymph node was reported in the PET/CT studies. For each FDG-PET/CT study, the SUVmax values of the most FDG avid pelvic lymph node were measured [9]. Treatment Patients without distant metastasis were treated with a combination of 3D conformal external beam RT (3DCRT) with concurrent weekly 40 mg/m2 cisplatin and high-dose-rate brachytherapy (HDR BRT) as previously described [12]. A total of 50.4 Gy external RT (1.8 Gy per fraction, daily, Monday through Friday) was delivered using 18-MV photons. The para-aortic region was also included in patients with FDG uptake in para-aortic lymph nodes. Para-aortic fields were treated with 45 Gy in 1.8-Gy fraction doses. In patients with enlarged lymph nodes, an additional 9-Gy boost dose was given. Three-dimensional brachytherapy planning was performed using 7 Gy per fraction prescribed to the target minimum, given in four fractions. Clinical follow-up Clinical follow-up of patients was performed every 3 months for 2 years, then every 6 months up to 5 years and annually thereafter. At least 3 months after the completion of treatment, all patients received PET/CT scans, and other radiological images were taken in suspected cases. A nuclear medicine physician noted the response at post-treatment PET/CT. Biopsy was not performed before 6 months of completion of chemoradiotherapy (ChRT). Failure was defined as biopsy-proven recurrence or documented progression of disease in serial-imaging studies [13]. Failure patterns were determined by follow-up imaging studies and were divided into four groups: none, isolated local failure (central pelvis and/or pelvic lymph nodes), distant failure (including para-aortic and supraclavicular lymph nodes), and combined local plus distant failure. Statistical analysis The time to event was calculated as the time interval from the date of diagnosis to the date of first finding on clinical or imaging examination that suggested disease recurrence. Pelvic disease recurrence was defined as disease in the cervical tumor, pelvic lymph nodes, or both. Disease-free survival (DFS) and OS rates were calculated using the Kaplan–Meier method. Correlations between parameters were calculated using the Pearson test. Multivariate analyses were performed using the Cox proportional hazards model. Clinicopathological factors and follow-up data from our cervical cancer database were analyzed for correlations with SUVmax. p-Values b 0.05 were considered statistically significant. Results Patient characteristics and outcomes Patient characteristics are presented in Table 1A. More than half of the patients had FIGO stage IIB disease, and most patients had squamous cell carcinoma. All patients were treated with concurrent ChRT; 78 patients (84%) completed at least four cycles of chemotherapy (ChT), 9 (10%) completed three cycles, and 6 (6%) received two cycles of ChT during RT. The median follow-up for all patients and surviving patients was 29 months (range, 3–79 months) and 38 months (range, 6–79 months), respectively. Of the 93 patients in the study cohort, 40 (43%) developed
42
C. Onal et al. / Gynecologic Oncology 137 (2015) 40–46
Table 1 (A) Patient characteristics. (B) Pelvic lymph node SUVmax stratification according to risk factors.
A
A Characteristics
Number of patients
Age, median (range), years FIGO stage IB2 IIA IIB IIIA IIIB IVA Histology Squamous cell carcinoma Adenocarcinoma Lymph node metastasis Pelvic Pelvic and para-aortic
58 (30–89)
Percent (%)
9 2 49 10 21 2
10 2 52 11 23 2
87 6
93 7
70 23
75 25
B
SD Stage bIIB ≥IIB Lymph node metastasis Pelvic Pelvic and para-aortic Post-therapy metabolic response CR PR/PD
SUVmax ≥ 7.5 n (%)
60 ± 11 18.0 ± 6.9 5.3 ± 1.7 1.4 ± 0.5 4.0 ± 1.0
61 ± 12 21.4 ± 8.8 5.8 ± 1.9 3.0 ± 1.2 7.6 ± 2.6
7 (8) 39 (42)
4 (4) 43 (46)
0.25
40 (43) 6 (7)
30 (32) 17 (18)
0.009
42 (45) 6 (7)
30 (32) 15 (16)
0.03
p-Value
0.91 0.04 0.13 b0.001 b0.001
local, locoregional, or distant failure. Of these, 23 (24%) developed distant metastases, 10 (11%) had pelvic recurrence and 7 (8%) developed both pelvic and distant failure. Of 17 patients with pelvic recurrence, 7 patients had central recurrence, 2 patients had lymphatic recurrence and 8 had both central and lymphatic recurrence. Of 10 patients with lymphatic recurrence, 9 recurred at the high SUVmax lymph node region, and one recurred at another lymph node area. At the time of the last follow-up, 47 patients (51%) were alive (9 [10%] with disease), and 46 patients (49%) had died. Of these latter patients, 37 (39%) died due to disease and 9 (10%) died from other causes.
Pelvic lymph node SUVmax
Age, mean ± SD (years) Primary tumor SUVmax, mean ± SD Primary tumor size, mean ± SD (cm) Pelvic lymph node size, mean ± SD (cm) Para-aortic lymph node SUVmax, mean ±
SUVmax b 7.5 n (%)
Pelvic lymph node SUVmax
Characteristics
B
C
FDG PET/CT findings The mean SUVmax for primary cervical tumor, pelvic lymph node, and para-aortic lymph node was 19.7 ± 8.0, 8.4 ± 4.3, and 6.7 ± 2.8, respectively. The distribution of the pelvic lymph node SUVmax is shown in Fig. 1A. There was a weak correlation between the primary tumor SUVmax and pelvic lymph node SUVmax (Pearson correlation coefficient [r] = 0.357; p b 0.001) (Fig. 1B). The average size of the primary tumor and FDG-avid pelvic lymph nodes was 5.5 ± 1.8 cm and 2.2 ± 1.2 cm, respectively. Pelvic lymph node size and SUVmax of pelvic lymph nodes were significantly correlated (r = 0.859; p b 0.001) (Fig. 1C). The average number of pelvic lymph node metastasis was 2 (range 1–8). There was no significant correlation between the number of pelvic lymph nodes and para-aortic lymph node metastasis. PET/CT images were obtained for all patients pre- and posttreatment. Post-treatment PET/CT images were taken within a median
Fig. 1. (A) Histogram of the maximum standardized uptake value (SUVmax) distribution of pelvic lymph nodes. Regression plots of (B) the SUVmax of primary cervical tumors versus the SUVmax of pelvic lymph nodes (r = 0.357; p b 0.001) and (C) maximum pelvic lymph node size versus the SUVmax of pelvic lymph nodes (r = 0.859; p b 0.001).
of 3.4 months (range, 3.0–8.9 months) after the completion of ChRT. At post-treatment PET/CT, 72 patients (77%) exhibited a complete response, 20 (22%) showed a partial response, and 1 (1%) showed progressive disease. The mean SUVmax of pelvic lymph nodes in the patients with locoregional or distant failure was significantly higher than that in
C. Onal et al. / Gynecologic Oncology 137 (2015) 40–46
43
the patients without any evidence of recurrence (9.6 ± 4.8 vs. 7.2 ± 3.2; p = 0.005). Although the pretreatment pelvic lymph node SUVmax in patients with a complete response trended lower than the pretreatment primary tumor SUVmax in patients with partial response or progressive disease, the difference was not statistically significant (8.0 ± 4.1 vs. 9.6 ± 4.8; p = 0.12). Prognostic factors Patients with pelvic and para-aortic lymph node metastases had significantly worse OS compared to patients with pelvic-only metastasis (p = 0.04) (Fig. 2A). Although the DFS rate trended lower in patients with pelvic and para-aortic lymph node metastases compared to patients with pelvic only metastasis, the difference fell short of significance (p = 0.07) (Fig. 2B). Patients with pelvic and para-aortic lymph node metastases had significantly higher SUVmax values for both primary tumor (23.4 ± 9.2 vs. 18.5 ± 7.3; p = 0.01) and pelvic lymph nodes (11.4 ± 4.6 vs. 7.4 ± 3.8; p = 0.001) compared to patients with only pelvic lymph node metastasis. Additionally, metastatic lymph node size was greater in patients with pelvic and para-aortic lymph node metastases compared to patients with pelvic-only lymph node metastasis (3.0 ± 1.4 cm vs. 1.9 ± 1.0 cm; p = 0.002).
Fig. 3. ROC curve analysis in recurrence-prediction according to the SUVmax of pelvic lymph nodes.
SUVmax analysis The relationship between primary tumor FDG uptake and survival was evaluated based on the cutoff value determined using receiver
A
P
P + PA PA
B
P
P + PA
Fig. 2. (A) OS and (B) DFS curves according to lymph node metastasis location. P, pelvic lymph node; P + PA, pelvic and para-aortic lymph nodes.
operating characteristic (ROC) curve analysis. Fig. 3 shows the ROC curve analysis of the pelvic lymph node SUVmax with respect to DFS. The area under the curve was 0.732 (p b 0.001; 95% confidence interval, 0.626–0.838), and the cutoff value of SUVmax in the present study was determined to be 7.5. As shown in Fig. 4A, OS (p = 0.04) rates were significantly higher in patients with SUVmax b 7.5 compared with those with SUVmax ≥ 7.5 for all groups, while there was a borderline significance (p = 0.08) for patients with pelvic lymph node metastasis only (Fig. 4B). Elevated pelvic lymph node SUVmax was found to be predictive of worse DFS for all patients (p = 0.02) (Fig. 4C) and for patients with pelvic lymph node metastasis (p = 0.04) (Fig. 4D). Patients with a pelvic lymph node SUVmax ≥ 7.5 had significantly higher SUVmax for both primary tumor and para-aortic lymph nodes, and larger metastatic pelvic lymph nodes compared to patients with a pelvic lymph node SUVmax b 7.5 (Table 1B). Post-therapy complete response rates were significantly higher in patients with a pelvic lymph node SUVmax b 7.5 compared to patients with a pelvic lymph node SUVmax ≥ 7.5. The number of patients with no evidence of disease was higher in the group with SUVmax b 7.5 (26 patients [28%]) compared with the group with SUVmax ≥ 7.5 ((10 patients [11%]); p b 0.001). Eleven patients (12%) in the group with SUVmax b 7.5 died, whereas 28 patients (30%) died in the group with SUVmax ≥ 7.5. In a multivariate analysis, pelvic lymph node SUVmax and posttherapy metabolic response were significant prognostic factors for both OS and DFS (Table 2A). Additionally, FIGO stage was borderline significant for OS and DFS. However there were no significant prognostic factors predicting OS and DFS for patients with pelvic lymph node metastasis only (Table 2B). Discussion The present study investigated the prognostic importance of the pelvic lymph node SUVmax in patients with cervical cancer. Patients with a higher pelvic lymph node SUVmax had higher rates of disease recurrence and worse survival. Additionally, we found that patients with a higher pelvic lymph node SUVmax had significantly larger tumors, higher incidences of para-aortic lymph node metastasis, worse response rates, and a borderline significant rate of advanced disease. Furthermore, we demonstrated that pelvic lymph node SUVmax and post-therapy metabolic response were significant prognostic factors for both OS and DFS.
44
C. Onal et al. / Gynecologic Oncology 137 (2015) 40–46
A
SUVmaax < 7..5
B
SUVmax < 7.5 7
SUVmax ≥ 7.5
7.5 SU UVmax ≥7 m
C
SUVmax < 7.5 7
SUVmax ≥ 7 7.5
D
SUVmaxx < 7.5
SUVmax ≥ 7 7.5
Fig. 4. Overall survival curves of patients with pelvic lymph node SUVmax of less than 7.5 and 7.5 or greater (A) for all patients and (B) those with pelvic lymph node metastasis only. Disease-free survival curves of patients with pelvic lymph node SUVmax of less than 7.5 and 7.5 or greater (C) for all patients and (D) those with pelvic lymph node metastasis only.
Table 2 Multivariate analysis of prognostic factors for DFS and OS (A) for all patients and (B) for patients with pelvic lymph node metastasis. Variables A Overall survival Pelvic lymph node SUVmax Stage Post-therapy response Lymph node metastasis Disease-free survival Pelvic lymph node SUVmax Stage Post-therapy response Lymph node metastasis
Risk factors
HR (95% CI)
p-Value
2.10 (2.03–2.17) 0.005 bIIB vs. ≥IIB CR vs. PR/PD Pelvic vs. pelvic and para-aortic
3.67 (0.88–5.29) 0.07 2.11 (1.15–3.89) 0.02 2.14 (1.17–3.81) 0.02
2.10 (2.04–2.18) 0.003 bIIB vs. ≥IIB CR vs. PR/PD Pelvic vs. pelvic and para-aortic
B Overall survival Pelvic lymph node SUVmax Post-therapy response CR vs. PR/PD Disease-free survival Pelvic lymph node SUVmax Post-therapy response CR vs. PR/PD
2.96 (0.92–9.53) 0.07 2.44 (1.40–4.25) 0.02 2.29 (1.27–2.39) 0.004
1.59 (0.71–3.58) 0.26 1.84 (0.75–4.51) 0.18 1.48 (0.70–3.14) 0.31 2.07 (0.91–4.74) 0.08
Abbreviations: DFS, disease-free survival; OS, overall survival.
Clinical stage, tumor volume, and nodal metastasis are the strongest prognostic factors in patients with locally advanced cervical cancer [2,14]. Although lymph node status does not contribute to the FIGO staging system for cervical carcinoma, it has been well demonstrated that lymph node metastasis decreases 5-year OS rates by 35–40% compared to patients without lymph node metastasis [8]. The 5-year OS rates for patients with pelvic and para-aortic lymph node metastases for all stages were 20–60% [15]. For this reason, accurate evaluation of lymph nodes is essential for predicting the prognosis of the disease. Moreover, although pelvic lymphatics were routinely irradiated in locally advanced disease, additional para-aortic RT fields are required in cases of para-aortic lymph node metastasis. The usefulness of pretreatment FDG PET in evaluating the extent of disease in patients with newly diagnosed carcinoma of the cervix has been well established. In these patients, the sensitivity and specificity of PET/CT for detection of pelvic lymph nodes metastasis were shown to be 79% and 99%, respectively, which are higher than those of CT and MRI [16]. Thus, PET/CT is accepted as an important tool for assessing lymph nodes in cervical cancer. The correlation between SUVmax and lymph node metastasis is not well demonstrated. In a study by Grigsby et al. [17], the presence of lymph node FDG uptake correlated most significantly with DFS. Kidd et al. [6] found a significantly higher SUVmax in patients with lymph node metastasis (157 patients [55%]) compared with those with no metastasis (130 patients [45%]), similar to our findings. We previously found a significant correlation between lymph node metastasis and primary tumor SUVmax in 149 patients with cervical
C. Onal et al. / Gynecologic Oncology 137 (2015) 40–46
cancer treated with definitive ChRT [7]. In that study, mean SUVmax was 19.7 ± 8.2 for patients with pelvic and/or para-aortic lymph node metastasis, and 16.4 ± 8.2 for patients without lymph node metastasis (p = 0.01). Although the primary tumor SUVmax has shown prognostic significance, the lymph node SUVmax has not been well studied. Yen et al. [10] demonstrated that a SUVmax ≥ 3.3 in para-aortic lymph nodes was significantly associated with OS (p = 0.01) and marginally associated with recurrence-free survival (RFS) (p = 0.08); however, no significant association has been established between pelvic lymph node SUVmax and OS or RFS. Interestingly, a significant correlation was observed in only 14 of 70 cases of para-aortic lymph node involvement. Kidd et al. [9] showed that patients with a pelvic lymph node SUVmax of ≥ 4.3 had worse DFS (p = 0.02) and OS (p = 0.04). In the present study, we used a SUVmax of 7.5, which was obtained from ROC curve analysis. We found a statistically significant association between pelvic lymph node SUVmax (b7.5 vs. ≥7.5) in pretreatment FDG PET and OS and DFS in patients with newly diagnosed carcinoma of the cervix. However, since our primary tumor and pelvic lymph node mean SUVmax were higher than those reported in previous studies, the cutoff value for pelvic lymph node SUVmax in the current study was also higher. Previous research has shown that lymph node status, as revealed by FDG-PET, is an important prognostic factor in patients with cervical cancer, and the results of the current study further demonstrate that the degree of FDG uptake by pelvic lymph nodes is also significant for treatment response, OS, and DFS. The pelvic lymph node SUVmax was significantly correlated with pelvic lymph node size in the current study, a result at odds with previous studies [9,10]. This finding may be related to the higher rate of extensive-stage disease (88%) in our study compared to previous studies, which would translate into a higher risk of lymph node metastasis in our study cohort. This finding demonstrates that both lymph node size and glucose metabolism in metastatic lymph nodes have a significant impact on treatment outcome. Although our previous research has indicated that a high cervical SUVmax is predictive of persistent disease after treatment, the findings of the current study demonstrate that an elevated pelvic lymph node SUVmax is predictive of an increased risk of persistent disease. In this study, we already found that patients with central or lymphatic recurrence had significantly higher pelvic lymph node SUVmax compared to patients with distant metastasis (14.1 ± 2.6 vs. 9.4 ± 4.7; p = 0.002). Thus, knowing at diagnosis that cervical cancer patients with highly FDG-avid lymph nodes are at increased risk of disease recurrence in the pelvic lymph nodes is valuable for making decisions regarding prognosis, treatment planning, and follow-up. Our study has some limitations. First, it is a retrospective study with limited number of patients, restricting the generalizability of the results. Second, the partial volume effects of PET/CT in detecting small lymph nodes may underestimate metastatic lymph nodes. Additionally, to minimize the complications of ChRT and prevent potential delays in starting ChRT, we did not perform histopathological verification of pelvic lymph nodes. Since it has been demonstrated that PET and PET/CT are reliable methods for detecting lymph node metastasis, we only evaluated patients with PET-positive pelvic and/or para-aortic lymph nodes [18]. Although PET only, which has limited resolution, has been used in previous studies, in the current study, we used PET/CT in order to increase the resolution of images and accuracy of detecting small lymph node metastases. Third, the inclusion of patients with various stages of disease may have introduced bias into the study, which may have affected the determination of the prevalence of lymph node metastases as well as the size of the involved lymph nodes. Our study offers some unique and significant findings, and differs from previous report. Recently, another group of researchers reported that the pelvic or para-aortic node SUVmax in patients with cervical carcinoma was predictive of disease control and survival in a study population of 83 patients [9]. In the current study, we did not include
45
patients with supraclavicular lymph node metastasis, which has a worse outcome. However, in the study by Kidd et al., 10 patients with supraclavicular metastasis were included in the study cohort, which may have influenced the results. In another study, Yen et al. [10] demonstrated the prognostic significance of the para-aortic lymph node SUVmax in 14 patients, without demonstrating the prognostic importance of the pelvic lymph node SUVmax. Since the lymphatic spread in cervical cancer occurs mainly in an orderly fashion – from pelvic to para-aortic lymph nodes – these may be conflicting results. In these studies, the authors evaluated lymph nodes by PET imaging; however, we prefer hybrid PET/CT imaging because of its improved resolution. Our findings, based on a larger and more homogenous patient population than those of previous studies and better evaluation of lymph nodes using PET/CT, clearly demonstrate that the pelvic lymph node SUVmax is strongly correlated with treatment response and survival. The patients with highly FDG-avid pelvic lymph nodes had higher risk of disease recurrence with worse survival, warranting higher boost radiation doses or additional adjuvant treatment strategies. Identification of these high-risk patients may permit the evaluation of the clinical benefits of additional treatments such as systemic ChT. A multiinstitutional clinical trial could confirm these findings together with other PET/CT-based prognostic factors. Conflict of interest statement The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References [1] Eifel PJ, Winter K, Morris M, Levenback C, Grigsby PW, Cooper J, Rotman M, Gershenson D, Mutch DG. Pelvic irradiation with concurrent chemotherapy versus pelvic and para-aortic irradiation for high-risk cervical cancer: an update of radiation therapy oncology group trial (RTOG) 90-01. J Clin Oncol 2004;22: 872–80. [2] Atahan IL, Onal C, Ozyar E, Yiliz F, Selek U, Kose F. Long-term outcome and prognostic factors in patients with cervical carcinoma: a retrospective study. Int J Gynecol Cancer 2007;17:833–42. [3] Tsai CS, Lai CH, Wang CC, Chang JT, Chang TC, Tseng CJ, Hong JH. The prognostic factors for patients with early cervical cancer treated by radical hysterectomy and postoperative radiotherapy. Gynecol Oncol 1999;75:328–33. [4] Wong TZ, Jones EL, Coleman RE. Positron emission tomography with 2-deoxy-2[(18)F]fluoro-d-glucose for evaluating local and distant disease in patients with cervical cancer. Mol Imaging Biol 2004;6:55–62. [5] Xue F, Lin LL, Dehdashti F, Miller TR, Siegel BA, Grigsby PW. F-18 fluorodeoxyglucose uptake in primary cervical cancer as an indicator of prognosis after radiation therapy. Gynecol Oncol 2006;101:147–51. [6] Kidd EA, Siegel BA, Dehdashti F, Grigsby PW. The standardized uptake value for F-18 fluorodeoxyglucose is a sensitive predictive biomarker for cervical cancer treatment response and survival. Cancer 2007;110:1738–44. [7] Onal C, Reyhan M, Parlak C, Guler OC, Oymak E. Prognostic value of pretreatment 18 F-fluorodeoxyglucose uptake in patients with cervical cancer treated with definitive chemoradiotherapy. Int J Gynecol Cancer 2013;23:1104–10. [8] Onal C, Reyhan M, Guler OC, Yapar AF. Treatment outcomes of patients with cervical cancer with complete metabolic responses after definitive chemoradiotherapy. Eur J Nucl Med Mol Imaging 2014;41:1336–42. [9] Kidd EA, Siegel BA, Dehdashti F, Grigsby PW. Pelvic lymph node F-18 fluorodeoxyglucose uptake as a prognostic biomarker in newly diagnosed patients with locally advanced cervical cancer. Cancer 2010;116:1469–75. [10] Yen TC, See LC, Lai CH, Tsai CS, Chao A, Hsueh S, Hong JH, Chang TC, Ng KK. Standardized uptake value in para-aortic lymph nodes is a significant prognostic factor in patients with primary advanced squamous cervical cancer. Eur J Nucl Med Mol Imaging 2008;35:493–501. [11] Onal C, Oymak E, Findikcioglu A, Reyhan M. Isolated mediastinal lymph node false positivity of [18F]-fluorodeoxyglucose-positron emission tomography/ computed tomography in patients with cervical cancer. Int J Gynecol Cancer 2013;23:337–42. [12] Onal C, Arslan G, Topkan E, Pehlivan B, Yavuz M, Oymak E, Yavuz A. Comparison of conventional and CT-based planning for intracavitary brachytherapy for cervical cancer: target volume coverage and organs at risk doses. J Exp Clin Cancer Res 2009;28:95. [13] Schwarz JK, Siegel BA, Dehdashti F, Grigsby PW. Association of posttherapy positron emission tomography with tumor response and survival in cervical carcinoma. JAMA 2007;298:2289–95. [14] Kidd EA, Siegel BA, Dehdashti F, Rader JS, Mutch DG, Powell MA, Grigsby PW. Lymph node staging by positron emission tomography in cervical cancer: relationship to prognosis. J Clin Oncol 2010;28:2108–13.
46
C. Onal et al. / Gynecologic Oncology 137 (2015) 40–46
[15] Parker K, Gallop-Evans E, Hanna L, Adams M. Five years' experience treating locally advanced cervical cancer with concurrent chemoradiotherapy and high-dose-rate brachytherapy: results from a single institution. Int J Radiat Oncol Biol Phys 2009;74:140–6. [16] Reinhardt MJ, Ehritt-Braun C, Vogelgesang D, Ihling C, Hogerle S, Mix M, Moser E, Krause TM. Metastatic lymph nodes in patients with cervical cancer: detection with MR imaging and FDG PET. Radiology 2001;218:776–82.
[17] Grigsby PW, Siegel BA, Dehdashti F, Rader J, Zoberi I. Posttherapy [ 18 F] fluorodeoxyglucose positron emission tomography in carcinoma of the cervix: response and outcome. J Clin Oncol 2004;22:2167–71. [18] Gouy S, Morice P, Narducci F, Uzan C, Gilmore J, Kolesnikov-Gauthier H, Querleu D, Haie-Meder C, Leblanc E. Nodal-staging surgery for locally advanced cervical cancer in the era of PET. Lancet Oncol 2012;13:e212–20.
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Prognostic value of 18F-fluorodeoxyglucose uptake in pelvic lymph nodes in patients with cervical cancer treated with definitive chemoradiotherapy Cem Onal a,⁎, Ozan C. Guler a, Mehmet Reyhan b, Ali Fuat Yapar b a b
Department of Radiation Oncology, Baskent University Faculty of Medicine, Adana, Turkey Department of Nuclear Medicine, Baskent University Faculty of Medicine, Ankara, Turkey
H I G H L I G H T S • The prognostic value of FDG uptake in regional lymph nodes for cervical cancer patients has been rarely studied. • Pelvic lymph node SUVmax, post-therapy metabolic response and para-aortic lymph node metastases were significant prognostic factors for survival. • High-risk patients may benefits from additional treatments such as systemic chemotherapy.
a r t i c l e
i n f o
Article history: Received 24 October 2014 Accepted 23 January 2015 Available online 29 January 2015 Keywords: Cervical cancer Chemoradiotherapy Lymph node metastasis Positron emission tomography
a b s t r a c t Purpose. To evaluate the prognostic significance of the maximum standardized uptake (SUVmax) value for pelvic lymph nodes in patients with cervical cancer and its impact on treatment response, disease control, and survival. Methods. Ninety-three patients with pelvic or para-aortic metastasis detected by PET/CT and treated with definitive chemoradiotherapy were evaluated. The impact of pelvic lymph node SUVmax on prognostic factors and treatment outcomes was assessed. Results. The size and SUVmax of pelvic lymph nodes were significantly correlated (r = 0.859; p b 0.001). Patients with pelvic and para-aortic lymph node metastases had significantly higher SUVmax values for both primary tumor (23.4 ± 9.2 vs. 18.5 ± 7.3; p = 0.01) and pelvic lymph nodes (11.4 ± 4.6 vs. 7.4 ± 3.8; p = 0.001). Patients with pelvic lymph node SUVmax ≥ 7.5 had significantly higher primary tumor SUVmax, larger pelvic lymph nodes, higher rates of para-aortic lymph node metastasis, and lower post-therapy complete response rates. Overall survival (OS) and disease-free survival (DFS) rates were significantly higher in patients with SUVmax b 7.5 compared to patients with SUVmax ≥ 7.5. In a multivariate analysis, pelvic lymph node SUVmax and post-therapy metabolic response were significant prognostic factors for both OS and DFS for all patients, but no significant prognostic factors were found in pelvic lymph node metastasis only. Conclusions. Patients with highly FDG-avid pelvic lymph nodes have a higher risk of disease recurrence with worse survival. Identification of these patients may assist in the evaluation of the clinical benefits of additional treatments. © 2015 Elsevier Inc. All rights reserved.
Introduction Cervical cancer is the second-most frequently diagnosed gynecological cancer worldwide. Almost half of such patients present with advanced disease at the time of initial diagnosis. Some prognostic factors have been associated with clinical outcome, including young ⁎ Corresponding author at: Baskent University Faculty of Medicine, Adana Research and Treatment Centre, Department of Radiation Oncology, 01120 Adana, Turkey. Fax: +90 322 3444445. E-mail address: [email protected] (C. Onal).
http://dx.doi.org/10.1016/j.ygyno.2015.01.542 0090-8258/© 2015 Elsevier Inc. All rights reserved.
age, pretreatment stage, and lymph node status [1,2]. Pelvic and/or para-aortic lymph node metastases have a strong impact on the prognosis and management of treatment in patients with cervical cancer. The lymphatic spread of cervical cancer occurs mainly in an orderly fashion, from pelvic to para-aortic lymph nodes to supraclavicular and mediastinal lymph nodes in advanced stages. Additionally, pelvic lymph node metastasis increases the risk of para-aortic lymph node metastasis [3]. 18 F-fluorodeoxyglucose (FDG) positron emission tomography (PET) integrated with computed tomography (PET/CT) incorporates metabolic tumor function with anatomic localization. PET/CT has become an increasingly important component of tumor staging by virtue of its
C. Onal et al. / Gynecologic Oncology 137 (2015) 40–46
ability to identify lymph node involvement, distant disease, and recurrent disease in patients with cervical cancer [4]. Some studies have also assessed the relationship between FDG uptake and outcome in cervical cancer [5,6]. Previously, we found that the maximum standardized uptake value (SUVmax) for the primary tumor was correlated with increased tumor size and lymph node metastasis rates, persistence of tumor after treatment, and worse overall survival—even after a complete metabolic response [7,8]. However, the prognostic value of FDG uptake in regional lymph nodes for cervical cancer patients has been rarely studied, and the predictive value of FDG uptake in pelvic lymph nodes remains unclear [9,10]. In a study involving patients with cervical cancer, Yen et al. [10] found that an SUVmax N 3.3 in para-aortic lymph nodes was predictive of worse overall survival in 14 of 70 patients with para-aortic lymph node metastasis, but found no impact of the pelvic lymph node SUVmax on overall survival (OS). In another study, Kidd et al. [9] demonstrated that the pelvic lymph node SUVmax was a prognostic biomarker and predictor of treatment outcome in 83 patients with cervical cancer. Given these recent studies suggesting that the SUVmax of involved lymph nodes is predictive of disease control and OS, and our previous research demonstrating the prognostic value of the primary tumor SUVmax as determined by FDG-PET [8], we sought to evaluate the prognostic significance of the pelvic lymph node SUVmax in patients with cervical cancer. Additionally, we analyzed the impact of the pelvic lymph node SUVmax on treatment response, disease control, and survival. Materials and methods Patients The study population consisted of 183 patients with biopsy-proven cervical cancer treated with curative intent between November 2006 and August 2013; 93 patients (51%) had pelvic or para-aortic metastasis detected by PET/CT. Approval was obtained from the institutional review board for this retrospective outcome analysis. All patients underwent routine clinical staging, including recording of medical history, physical and gynecological examinations, blood chemistry tests, complete blood count, and magnetic resonance imaging (MRI) or CT of the abdomen and pelvis. The patients were staged according to the International Federation of Gynecology and Obstetrics (FIGO) staging system. All patients had also undergone PET/CT for initial diagnosis, staging, and radiotherapy (RT) planning. PET/CT technique The patients were imaged using a dedicated PET/CT system (Discovery-STE 8; General Electric Medical Systems, Milwaukee, WI) as previously described [11]. The patients fasted for at least 6 h before intravenous administration of 370 to 555 MBq (10–15 mCi) FDG. Preinjection blood glucose levels were measured to ensure that they were below 150 mg/dL. During the distribution phase, the patients laid supine in a quiet room. Combined image acquisition began 60 min after FDG injection. The patients were scanned on a flat-panel, carbon-fiber composite table insert. First, an unenhanced CT scan (5-mm slice thickness) from the base of the skull to the inferior border of the pelvis was acquired using a standardized protocol (140 kV and 80 mA). The subsequent PET scan was acquired in three-dimensional (3D) mode from the base of the skull to the inferior border of the pelvis (6 to 7 bed positions, 3 min per position) without repositioning the patient on the table. CT and PET images were acquired with the patient breathing shallowly. Attenuation was corrected using the CT images. Areas of FDG uptake were categorized as malignant based on location, intensity, shape, size, and visual correlation with CT images to differentiate physiologic uptake from pathologic uptake.
41
A lymph node was considered PET-positive if its FDG uptake was greater than blood pool activity or surrounding background tissues, depending on the size of the node. The maximum long axis of a pelvic lymph node with FDG uptake was recorded at co-registered PET/CT images. Patients were eligible for this study if at least one FDG-avid pelvic lymph node was reported in the PET/CT studies. For each FDG-PET/CT study, the SUVmax values of the most FDG avid pelvic lymph node were measured [9]. Treatment Patients without distant metastasis were treated with a combination of 3D conformal external beam RT (3DCRT) with concurrent weekly 40 mg/m2 cisplatin and high-dose-rate brachytherapy (HDR BRT) as previously described [12]. A total of 50.4 Gy external RT (1.8 Gy per fraction, daily, Monday through Friday) was delivered using 18-MV photons. The para-aortic region was also included in patients with FDG uptake in para-aortic lymph nodes. Para-aortic fields were treated with 45 Gy in 1.8-Gy fraction doses. In patients with enlarged lymph nodes, an additional 9-Gy boost dose was given. Three-dimensional brachytherapy planning was performed using 7 Gy per fraction prescribed to the target minimum, given in four fractions. Clinical follow-up Clinical follow-up of patients was performed every 3 months for 2 years, then every 6 months up to 5 years and annually thereafter. At least 3 months after the completion of treatment, all patients received PET/CT scans, and other radiological images were taken in suspected cases. A nuclear medicine physician noted the response at post-treatment PET/CT. Biopsy was not performed before 6 months of completion of chemoradiotherapy (ChRT). Failure was defined as biopsy-proven recurrence or documented progression of disease in serial-imaging studies [13]. Failure patterns were determined by follow-up imaging studies and were divided into four groups: none, isolated local failure (central pelvis and/or pelvic lymph nodes), distant failure (including para-aortic and supraclavicular lymph nodes), and combined local plus distant failure. Statistical analysis The time to event was calculated as the time interval from the date of diagnosis to the date of first finding on clinical or imaging examination that suggested disease recurrence. Pelvic disease recurrence was defined as disease in the cervical tumor, pelvic lymph nodes, or both. Disease-free survival (DFS) and OS rates were calculated using the Kaplan–Meier method. Correlations between parameters were calculated using the Pearson test. Multivariate analyses were performed using the Cox proportional hazards model. Clinicopathological factors and follow-up data from our cervical cancer database were analyzed for correlations with SUVmax. p-Values b 0.05 were considered statistically significant. Results Patient characteristics and outcomes Patient characteristics are presented in Table 1A. More than half of the patients had FIGO stage IIB disease, and most patients had squamous cell carcinoma. All patients were treated with concurrent ChRT; 78 patients (84%) completed at least four cycles of chemotherapy (ChT), 9 (10%) completed three cycles, and 6 (6%) received two cycles of ChT during RT. The median follow-up for all patients and surviving patients was 29 months (range, 3–79 months) and 38 months (range, 6–79 months), respectively. Of the 93 patients in the study cohort, 40 (43%) developed
42
C. Onal et al. / Gynecologic Oncology 137 (2015) 40–46
Table 1 (A) Patient characteristics. (B) Pelvic lymph node SUVmax stratification according to risk factors.
A
A Characteristics
Number of patients
Age, median (range), years FIGO stage IB2 IIA IIB IIIA IIIB IVA Histology Squamous cell carcinoma Adenocarcinoma Lymph node metastasis Pelvic Pelvic and para-aortic
58 (30–89)
Percent (%)
9 2 49 10 21 2
10 2 52 11 23 2
87 6
93 7
70 23
75 25
B
SD Stage bIIB ≥IIB Lymph node metastasis Pelvic Pelvic and para-aortic Post-therapy metabolic response CR PR/PD
SUVmax ≥ 7.5 n (%)
60 ± 11 18.0 ± 6.9 5.3 ± 1.7 1.4 ± 0.5 4.0 ± 1.0
61 ± 12 21.4 ± 8.8 5.8 ± 1.9 3.0 ± 1.2 7.6 ± 2.6
7 (8) 39 (42)
4 (4) 43 (46)
0.25
40 (43) 6 (7)
30 (32) 17 (18)
0.009
42 (45) 6 (7)
30 (32) 15 (16)
0.03
p-Value
0.91 0.04 0.13 b0.001 b0.001
local, locoregional, or distant failure. Of these, 23 (24%) developed distant metastases, 10 (11%) had pelvic recurrence and 7 (8%) developed both pelvic and distant failure. Of 17 patients with pelvic recurrence, 7 patients had central recurrence, 2 patients had lymphatic recurrence and 8 had both central and lymphatic recurrence. Of 10 patients with lymphatic recurrence, 9 recurred at the high SUVmax lymph node region, and one recurred at another lymph node area. At the time of the last follow-up, 47 patients (51%) were alive (9 [10%] with disease), and 46 patients (49%) had died. Of these latter patients, 37 (39%) died due to disease and 9 (10%) died from other causes.
Pelvic lymph node SUVmax
Age, mean ± SD (years) Primary tumor SUVmax, mean ± SD Primary tumor size, mean ± SD (cm) Pelvic lymph node size, mean ± SD (cm) Para-aortic lymph node SUVmax, mean ±
SUVmax b 7.5 n (%)
Pelvic lymph node SUVmax
Characteristics
B
C
FDG PET/CT findings The mean SUVmax for primary cervical tumor, pelvic lymph node, and para-aortic lymph node was 19.7 ± 8.0, 8.4 ± 4.3, and 6.7 ± 2.8, respectively. The distribution of the pelvic lymph node SUVmax is shown in Fig. 1A. There was a weak correlation between the primary tumor SUVmax and pelvic lymph node SUVmax (Pearson correlation coefficient [r] = 0.357; p b 0.001) (Fig. 1B). The average size of the primary tumor and FDG-avid pelvic lymph nodes was 5.5 ± 1.8 cm and 2.2 ± 1.2 cm, respectively. Pelvic lymph node size and SUVmax of pelvic lymph nodes were significantly correlated (r = 0.859; p b 0.001) (Fig. 1C). The average number of pelvic lymph node metastasis was 2 (range 1–8). There was no significant correlation between the number of pelvic lymph nodes and para-aortic lymph node metastasis. PET/CT images were obtained for all patients pre- and posttreatment. Post-treatment PET/CT images were taken within a median
Fig. 1. (A) Histogram of the maximum standardized uptake value (SUVmax) distribution of pelvic lymph nodes. Regression plots of (B) the SUVmax of primary cervical tumors versus the SUVmax of pelvic lymph nodes (r = 0.357; p b 0.001) and (C) maximum pelvic lymph node size versus the SUVmax of pelvic lymph nodes (r = 0.859; p b 0.001).
of 3.4 months (range, 3.0–8.9 months) after the completion of ChRT. At post-treatment PET/CT, 72 patients (77%) exhibited a complete response, 20 (22%) showed a partial response, and 1 (1%) showed progressive disease. The mean SUVmax of pelvic lymph nodes in the patients with locoregional or distant failure was significantly higher than that in
C. Onal et al. / Gynecologic Oncology 137 (2015) 40–46
43
the patients without any evidence of recurrence (9.6 ± 4.8 vs. 7.2 ± 3.2; p = 0.005). Although the pretreatment pelvic lymph node SUVmax in patients with a complete response trended lower than the pretreatment primary tumor SUVmax in patients with partial response or progressive disease, the difference was not statistically significant (8.0 ± 4.1 vs. 9.6 ± 4.8; p = 0.12). Prognostic factors Patients with pelvic and para-aortic lymph node metastases had significantly worse OS compared to patients with pelvic-only metastasis (p = 0.04) (Fig. 2A). Although the DFS rate trended lower in patients with pelvic and para-aortic lymph node metastases compared to patients with pelvic only metastasis, the difference fell short of significance (p = 0.07) (Fig. 2B). Patients with pelvic and para-aortic lymph node metastases had significantly higher SUVmax values for both primary tumor (23.4 ± 9.2 vs. 18.5 ± 7.3; p = 0.01) and pelvic lymph nodes (11.4 ± 4.6 vs. 7.4 ± 3.8; p = 0.001) compared to patients with only pelvic lymph node metastasis. Additionally, metastatic lymph node size was greater in patients with pelvic and para-aortic lymph node metastases compared to patients with pelvic-only lymph node metastasis (3.0 ± 1.4 cm vs. 1.9 ± 1.0 cm; p = 0.002).
Fig. 3. ROC curve analysis in recurrence-prediction according to the SUVmax of pelvic lymph nodes.
SUVmax analysis The relationship between primary tumor FDG uptake and survival was evaluated based on the cutoff value determined using receiver
A
P
P + PA PA
B
P
P + PA
Fig. 2. (A) OS and (B) DFS curves according to lymph node metastasis location. P, pelvic lymph node; P + PA, pelvic and para-aortic lymph nodes.
operating characteristic (ROC) curve analysis. Fig. 3 shows the ROC curve analysis of the pelvic lymph node SUVmax with respect to DFS. The area under the curve was 0.732 (p b 0.001; 95% confidence interval, 0.626–0.838), and the cutoff value of SUVmax in the present study was determined to be 7.5. As shown in Fig. 4A, OS (p = 0.04) rates were significantly higher in patients with SUVmax b 7.5 compared with those with SUVmax ≥ 7.5 for all groups, while there was a borderline significance (p = 0.08) for patients with pelvic lymph node metastasis only (Fig. 4B). Elevated pelvic lymph node SUVmax was found to be predictive of worse DFS for all patients (p = 0.02) (Fig. 4C) and for patients with pelvic lymph node metastasis (p = 0.04) (Fig. 4D). Patients with a pelvic lymph node SUVmax ≥ 7.5 had significantly higher SUVmax for both primary tumor and para-aortic lymph nodes, and larger metastatic pelvic lymph nodes compared to patients with a pelvic lymph node SUVmax b 7.5 (Table 1B). Post-therapy complete response rates were significantly higher in patients with a pelvic lymph node SUVmax b 7.5 compared to patients with a pelvic lymph node SUVmax ≥ 7.5. The number of patients with no evidence of disease was higher in the group with SUVmax b 7.5 (26 patients [28%]) compared with the group with SUVmax ≥ 7.5 ((10 patients [11%]); p b 0.001). Eleven patients (12%) in the group with SUVmax b 7.5 died, whereas 28 patients (30%) died in the group with SUVmax ≥ 7.5. In a multivariate analysis, pelvic lymph node SUVmax and posttherapy metabolic response were significant prognostic factors for both OS and DFS (Table 2A). Additionally, FIGO stage was borderline significant for OS and DFS. However there were no significant prognostic factors predicting OS and DFS for patients with pelvic lymph node metastasis only (Table 2B). Discussion The present study investigated the prognostic importance of the pelvic lymph node SUVmax in patients with cervical cancer. Patients with a higher pelvic lymph node SUVmax had higher rates of disease recurrence and worse survival. Additionally, we found that patients with a higher pelvic lymph node SUVmax had significantly larger tumors, higher incidences of para-aortic lymph node metastasis, worse response rates, and a borderline significant rate of advanced disease. Furthermore, we demonstrated that pelvic lymph node SUVmax and post-therapy metabolic response were significant prognostic factors for both OS and DFS.
44
C. Onal et al. / Gynecologic Oncology 137 (2015) 40–46
A
SUVmaax < 7..5
B
SUVmax < 7.5 7
SUVmax ≥ 7.5
7.5 SU UVmax ≥7 m
C
SUVmax < 7.5 7
SUVmax ≥ 7 7.5
D
SUVmaxx < 7.5
SUVmax ≥ 7 7.5
Fig. 4. Overall survival curves of patients with pelvic lymph node SUVmax of less than 7.5 and 7.5 or greater (A) for all patients and (B) those with pelvic lymph node metastasis only. Disease-free survival curves of patients with pelvic lymph node SUVmax of less than 7.5 and 7.5 or greater (C) for all patients and (D) those with pelvic lymph node metastasis only.
Table 2 Multivariate analysis of prognostic factors for DFS and OS (A) for all patients and (B) for patients with pelvic lymph node metastasis. Variables A Overall survival Pelvic lymph node SUVmax Stage Post-therapy response Lymph node metastasis Disease-free survival Pelvic lymph node SUVmax Stage Post-therapy response Lymph node metastasis
Risk factors
HR (95% CI)
p-Value
2.10 (2.03–2.17) 0.005 bIIB vs. ≥IIB CR vs. PR/PD Pelvic vs. pelvic and para-aortic
3.67 (0.88–5.29) 0.07 2.11 (1.15–3.89) 0.02 2.14 (1.17–3.81) 0.02
2.10 (2.04–2.18) 0.003 bIIB vs. ≥IIB CR vs. PR/PD Pelvic vs. pelvic and para-aortic
B Overall survival Pelvic lymph node SUVmax Post-therapy response CR vs. PR/PD Disease-free survival Pelvic lymph node SUVmax Post-therapy response CR vs. PR/PD
2.96 (0.92–9.53) 0.07 2.44 (1.40–4.25) 0.02 2.29 (1.27–2.39) 0.004
1.59 (0.71–3.58) 0.26 1.84 (0.75–4.51) 0.18 1.48 (0.70–3.14) 0.31 2.07 (0.91–4.74) 0.08
Abbreviations: DFS, disease-free survival; OS, overall survival.
Clinical stage, tumor volume, and nodal metastasis are the strongest prognostic factors in patients with locally advanced cervical cancer [2,14]. Although lymph node status does not contribute to the FIGO staging system for cervical carcinoma, it has been well demonstrated that lymph node metastasis decreases 5-year OS rates by 35–40% compared to patients without lymph node metastasis [8]. The 5-year OS rates for patients with pelvic and para-aortic lymph node metastases for all stages were 20–60% [15]. For this reason, accurate evaluation of lymph nodes is essential for predicting the prognosis of the disease. Moreover, although pelvic lymphatics were routinely irradiated in locally advanced disease, additional para-aortic RT fields are required in cases of para-aortic lymph node metastasis. The usefulness of pretreatment FDG PET in evaluating the extent of disease in patients with newly diagnosed carcinoma of the cervix has been well established. In these patients, the sensitivity and specificity of PET/CT for detection of pelvic lymph nodes metastasis were shown to be 79% and 99%, respectively, which are higher than those of CT and MRI [16]. Thus, PET/CT is accepted as an important tool for assessing lymph nodes in cervical cancer. The correlation between SUVmax and lymph node metastasis is not well demonstrated. In a study by Grigsby et al. [17], the presence of lymph node FDG uptake correlated most significantly with DFS. Kidd et al. [6] found a significantly higher SUVmax in patients with lymph node metastasis (157 patients [55%]) compared with those with no metastasis (130 patients [45%]), similar to our findings. We previously found a significant correlation between lymph node metastasis and primary tumor SUVmax in 149 patients with cervical
C. Onal et al. / Gynecologic Oncology 137 (2015) 40–46
cancer treated with definitive ChRT [7]. In that study, mean SUVmax was 19.7 ± 8.2 for patients with pelvic and/or para-aortic lymph node metastasis, and 16.4 ± 8.2 for patients without lymph node metastasis (p = 0.01). Although the primary tumor SUVmax has shown prognostic significance, the lymph node SUVmax has not been well studied. Yen et al. [10] demonstrated that a SUVmax ≥ 3.3 in para-aortic lymph nodes was significantly associated with OS (p = 0.01) and marginally associated with recurrence-free survival (RFS) (p = 0.08); however, no significant association has been established between pelvic lymph node SUVmax and OS or RFS. Interestingly, a significant correlation was observed in only 14 of 70 cases of para-aortic lymph node involvement. Kidd et al. [9] showed that patients with a pelvic lymph node SUVmax of ≥ 4.3 had worse DFS (p = 0.02) and OS (p = 0.04). In the present study, we used a SUVmax of 7.5, which was obtained from ROC curve analysis. We found a statistically significant association between pelvic lymph node SUVmax (b7.5 vs. ≥7.5) in pretreatment FDG PET and OS and DFS in patients with newly diagnosed carcinoma of the cervix. However, since our primary tumor and pelvic lymph node mean SUVmax were higher than those reported in previous studies, the cutoff value for pelvic lymph node SUVmax in the current study was also higher. Previous research has shown that lymph node status, as revealed by FDG-PET, is an important prognostic factor in patients with cervical cancer, and the results of the current study further demonstrate that the degree of FDG uptake by pelvic lymph nodes is also significant for treatment response, OS, and DFS. The pelvic lymph node SUVmax was significantly correlated with pelvic lymph node size in the current study, a result at odds with previous studies [9,10]. This finding may be related to the higher rate of extensive-stage disease (88%) in our study compared to previous studies, which would translate into a higher risk of lymph node metastasis in our study cohort. This finding demonstrates that both lymph node size and glucose metabolism in metastatic lymph nodes have a significant impact on treatment outcome. Although our previous research has indicated that a high cervical SUVmax is predictive of persistent disease after treatment, the findings of the current study demonstrate that an elevated pelvic lymph node SUVmax is predictive of an increased risk of persistent disease. In this study, we already found that patients with central or lymphatic recurrence had significantly higher pelvic lymph node SUVmax compared to patients with distant metastasis (14.1 ± 2.6 vs. 9.4 ± 4.7; p = 0.002). Thus, knowing at diagnosis that cervical cancer patients with highly FDG-avid lymph nodes are at increased risk of disease recurrence in the pelvic lymph nodes is valuable for making decisions regarding prognosis, treatment planning, and follow-up. Our study has some limitations. First, it is a retrospective study with limited number of patients, restricting the generalizability of the results. Second, the partial volume effects of PET/CT in detecting small lymph nodes may underestimate metastatic lymph nodes. Additionally, to minimize the complications of ChRT and prevent potential delays in starting ChRT, we did not perform histopathological verification of pelvic lymph nodes. Since it has been demonstrated that PET and PET/CT are reliable methods for detecting lymph node metastasis, we only evaluated patients with PET-positive pelvic and/or para-aortic lymph nodes [18]. Although PET only, which has limited resolution, has been used in previous studies, in the current study, we used PET/CT in order to increase the resolution of images and accuracy of detecting small lymph node metastases. Third, the inclusion of patients with various stages of disease may have introduced bias into the study, which may have affected the determination of the prevalence of lymph node metastases as well as the size of the involved lymph nodes. Our study offers some unique and significant findings, and differs from previous report. Recently, another group of researchers reported that the pelvic or para-aortic node SUVmax in patients with cervical carcinoma was predictive of disease control and survival in a study population of 83 patients [9]. In the current study, we did not include
45
patients with supraclavicular lymph node metastasis, which has a worse outcome. However, in the study by Kidd et al., 10 patients with supraclavicular metastasis were included in the study cohort, which may have influenced the results. In another study, Yen et al. [10] demonstrated the prognostic significance of the para-aortic lymph node SUVmax in 14 patients, without demonstrating the prognostic importance of the pelvic lymph node SUVmax. Since the lymphatic spread in cervical cancer occurs mainly in an orderly fashion – from pelvic to para-aortic lymph nodes – these may be conflicting results. In these studies, the authors evaluated lymph nodes by PET imaging; however, we prefer hybrid PET/CT imaging because of its improved resolution. Our findings, based on a larger and more homogenous patient population than those of previous studies and better evaluation of lymph nodes using PET/CT, clearly demonstrate that the pelvic lymph node SUVmax is strongly correlated with treatment response and survival. The patients with highly FDG-avid pelvic lymph nodes had higher risk of disease recurrence with worse survival, warranting higher boost radiation doses or additional adjuvant treatment strategies. Identification of these high-risk patients may permit the evaluation of the clinical benefits of additional treatments such as systemic ChT. A multiinstitutional clinical trial could confirm these findings together with other PET/CT-based prognostic factors. Conflict of interest statement The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References [1] Eifel PJ, Winter K, Morris M, Levenback C, Grigsby PW, Cooper J, Rotman M, Gershenson D, Mutch DG. Pelvic irradiation with concurrent chemotherapy versus pelvic and para-aortic irradiation for high-risk cervical cancer: an update of radiation therapy oncology group trial (RTOG) 90-01. J Clin Oncol 2004;22: 872–80. [2] Atahan IL, Onal C, Ozyar E, Yiliz F, Selek U, Kose F. Long-term outcome and prognostic factors in patients with cervical carcinoma: a retrospective study. Int J Gynecol Cancer 2007;17:833–42. [3] Tsai CS, Lai CH, Wang CC, Chang JT, Chang TC, Tseng CJ, Hong JH. The prognostic factors for patients with early cervical cancer treated by radical hysterectomy and postoperative radiotherapy. Gynecol Oncol 1999;75:328–33. [4] Wong TZ, Jones EL, Coleman RE. Positron emission tomography with 2-deoxy-2[(18)F]fluoro-d-glucose for evaluating local and distant disease in patients with cervical cancer. Mol Imaging Biol 2004;6:55–62. [5] Xue F, Lin LL, Dehdashti F, Miller TR, Siegel BA, Grigsby PW. F-18 fluorodeoxyglucose uptake in primary cervical cancer as an indicator of prognosis after radiation therapy. Gynecol Oncol 2006;101:147–51. [6] Kidd EA, Siegel BA, Dehdashti F, Grigsby PW. The standardized uptake value for F-18 fluorodeoxyglucose is a sensitive predictive biomarker for cervical cancer treatment response and survival. Cancer 2007;110:1738–44. [7] Onal C, Reyhan M, Parlak C, Guler OC, Oymak E. Prognostic value of pretreatment 18 F-fluorodeoxyglucose uptake in patients with cervical cancer treated with definitive chemoradiotherapy. Int J Gynecol Cancer 2013;23:1104–10. [8] Onal C, Reyhan M, Guler OC, Yapar AF. Treatment outcomes of patients with cervical cancer with complete metabolic responses after definitive chemoradiotherapy. Eur J Nucl Med Mol Imaging 2014;41:1336–42. [9] Kidd EA, Siegel BA, Dehdashti F, Grigsby PW. Pelvic lymph node F-18 fluorodeoxyglucose uptake as a prognostic biomarker in newly diagnosed patients with locally advanced cervical cancer. Cancer 2010;116:1469–75. [10] Yen TC, See LC, Lai CH, Tsai CS, Chao A, Hsueh S, Hong JH, Chang TC, Ng KK. Standardized uptake value in para-aortic lymph nodes is a significant prognostic factor in patients with primary advanced squamous cervical cancer. Eur J Nucl Med Mol Imaging 2008;35:493–501. [11] Onal C, Oymak E, Findikcioglu A, Reyhan M. Isolated mediastinal lymph node false positivity of [18F]-fluorodeoxyglucose-positron emission tomography/ computed tomography in patients with cervical cancer. Int J Gynecol Cancer 2013;23:337–42. [12] Onal C, Arslan G, Topkan E, Pehlivan B, Yavuz M, Oymak E, Yavuz A. Comparison of conventional and CT-based planning for intracavitary brachytherapy for cervical cancer: target volume coverage and organs at risk doses. J Exp Clin Cancer Res 2009;28:95. [13] Schwarz JK, Siegel BA, Dehdashti F, Grigsby PW. Association of posttherapy positron emission tomography with tumor response and survival in cervical carcinoma. JAMA 2007;298:2289–95. [14] Kidd EA, Siegel BA, Dehdashti F, Rader JS, Mutch DG, Powell MA, Grigsby PW. Lymph node staging by positron emission tomography in cervical cancer: relationship to prognosis. J Clin Oncol 2010;28:2108–13.
46
C. Onal et al. / Gynecologic Oncology 137 (2015) 40–46
[15] Parker K, Gallop-Evans E, Hanna L, Adams M. Five years' experience treating locally advanced cervical cancer with concurrent chemoradiotherapy and high-dose-rate brachytherapy: results from a single institution. Int J Radiat Oncol Biol Phys 2009;74:140–6. [16] Reinhardt MJ, Ehritt-Braun C, Vogelgesang D, Ihling C, Hogerle S, Mix M, Moser E, Krause TM. Metastatic lymph nodes in patients with cervical cancer: detection with MR imaging and FDG PET. Radiology 2001;218:776–82.
[17] Grigsby PW, Siegel BA, Dehdashti F, Rader J, Zoberi I. Posttherapy [ 18 F] fluorodeoxyglucose positron emission tomography in carcinoma of the cervix: response and outcome. J Clin Oncol 2004;22:2167–71. [18] Gouy S, Morice P, Narducci F, Uzan C, Gilmore J, Kolesnikov-Gauthier H, Querleu D, Haie-Meder C, Leblanc E. Nodal-staging surgery for locally advanced cervical cancer in the era of PET. Lancet Oncol 2012;13:e212–20.