Ann Nucl Med (2013) 27:37–45 DOI 10.1007/s12149-012-0659-3

ORIGINAL ARTICLE

Early response of patients undergoing concurrent chemoradiotherapy for cervical cancer: a comparison of PET/CT and MRI Jeong Eun Lee • Seung Jae Huh • Heerim Nam Sang Gyu Ju

•

Received: 24 June 2012 / Accepted: 17 September 2012 / Published online: 9 October 2012 Ó The Japanese Society of Nuclear Medicine 2012

Abstract Objective To investigate the efficacy of positron emission tomography/computed tomography (PET/CT) and magnetic resonance imaging (MRI) for early response evaluation of cervical cancer patients undergoing concurrent chemoradiotherapy (CCRT). Methods Fifty-two patients were prospectively enrolled in the study. The pathologic findings were squamous cell carcinoma in 47 patients and adenocarcinoma in 5 patients. All patients underwent PET/CT and MRI scans before, during and within 1 month after completion of CCRT. The percent change in tumor volume during and after CCRT based on PET/CT and MRI images was compared. Results There were significant differences (p \ 0.001) between the initial tumor volume and tumor volume during and after CCRT as measured by both PET/CT and MRI. During CCRT, the percent volume reduction based on PET/ CT images was significantly greater than the percent volume reduction calculated from MRI images (p = 0.024). However, after the completion of CCRT, no significant differences were found in volume reduction as calculated based on PET/CT versus MRI images (p = 0.289). The percent volume reduction of adenocarcinomas was significantly smaller than that of squamous cell carcinomas based on both PET/ CT (p = 0.041) and MRI images (p \ 0.001).

J. E. Lee Department of Radiation Oncology, Kyungpook National University School of Medicine, Daegu, Korea S. J. Huh (&)  H. Nam  S. G. Ju Department of Radiation Oncology, Samsung Medical Center, Sungkyunkwan University, School of Medicine, Irwon-dong 50, Gangnam-gu, Seoul 135-710, Korea e-mail: [email protected]

Conclusions Significant decreases in tumor volume were observed during and after CCRT in patients with cervical cancer. Tumor volume reduction on PET/CT images was greater than that on MRI images during CCRT. We suggest that early PET/CT as well as MRI scans could be taken during CCRT to evaluate tumor response and allow personalized treatment of cervical cancer. Keywords Cervical cancer  PET/CT  MRI  CCRT  Treatment response

Introduction Imaging modalities such as computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET) with 18F-fluorodeoxyglucose (FDG) are more effective tools for determining the extent of disease in cervical cancer. MRI is superior to CT for evaluating tumor size, stromal invasion, and local and regional extents of disease [1]. In contrast to CT and MRI, PET is used to assess the metabolic activity of the tumor. There is a strong correlation between tumor size measured by PET and pathologic tumor measurements in cervical cancer [2]. PET has been used for the pretreatment evaluation and in the routine surveillance of cervical cancer patients after treatment [3]. In the post-treatment setting, PET obtained 3 months after treatment can be used to predict treatment response [4]. But, if the response of the tumor could be estimated during or just after concurrent chemoradiotherapy (CCRT), patients with poor response could be identified earlier and their treatment strategies altered accordingly. Lin et al. [5] reported that a significant reduction in tumor volume, as measured by PET, occurred within the first 20 days of CCRT in cervical cancer patients.

123

38

Ann Nucl Med (2013) 27:37–45

We have previously reported that the volume regression rate during radiotherapy (RT) by MRI was a significant predictor of local control rate in cervical cancer patients [6]. There are several studies about the tumor volume correlation between MRI and PET/CT in cervical cancer patients at the time of diagnosis [7]. Our hypothesis is that if the tumor volume by MRI could predict the local control rate in cervical cancer patients, the tumor volume by PET/ CT also could be a possible predictor of local control rate. To the best of our knowledge, no previous studies have compared PET/CT and MRI for the quantitative evaluation of primary tumor volume in cervical cancer patients during CCRT. We therefore performed this prospective study to compare tumor volume reduction based on PET/CT and MRI measurements during and immediately after the completion of CCRT in patients with cervical cancer.

Methods and materials Patients and treatment Between January 2009 and September 2010, 52 women with biopsy-proven cervical cancer who required definitive CCRT were prospectively enrolled in this study. The study protocol was approved by the institutional review board of our institution. The median age of the patients was 54 years (range 28–85 years). There were 1 patient with FIGO stage Ib1, 7 patients with stage IIa, 36 with stage IIb, 5 with stage IIIb, and 3 with Stage IVa cervical cancer. Forty-seven patients had squamous cell carcinoma and 5 patients had adenocarcinoma. Details of the patients’ characteristics prior to treatment are provided in Table 1.

Table 1 Patient characteristics (n = 52) Characteristics

Number of patients (%)

FIGO stage IB1

1 (2)

IIA

7 (14)

IIB

36 (69)

IIIB

5 (10)

IVA

3 (6)

Histology Squamous cell carcinoma Adenocarcinoma Age (years) \39

47 (90) 5 (10) Median 54 (range 28–85) 5 (10)

40–59

30 (58)

C60

17 (32)

123

RT consisted of external beam RT (EBRT) and highdose-rate intracavitary brachytherapy with an iridium-192 source, which was initiated after an EBRT dose of 36.0–45.0 Gy (median dose 41.4 Gy) with midline shielding. EBRT was performed by applying 15 MV photons to the entire pelvis with the dose ranging from 41.4 to 55.8 Gy (median dose 50.4 Gy). The extended pelvic field, including the para-aortic lymph nodes, was targeted in two patients with para-aortic node involvement, with a dose of up to 45 Gy. Brachytherapy with Fletcher applicator (Nucletron, The Netherlands) was delivered three times a week with 4 Gy per insertion in 6 fractions to a total dose of 24 Gy. Three-dimensional image-guided brachytherapy planning (PLATO, version 14.3, Nucletron, The Netherlands) was performed. We prescribed 4 Gy to the volume encompassing 95 % of the isodose line, which covers clinical target volume (CTV). The CTV for brachytherapy included the gross tumor volume (GTV) on PET/CT images plus 1 cm margin toward the soft tissue mass boundaries of the uterus, cervix, and vagina and the entire cervix. All patients received CCRT in combination with weekly cisplatin at 40 mg/m2.

Image acquisition and volume measurements All patients underwent PET/CT and MRI scans before treatment (pre-Tx), during treatment (mid-Tx), and 1 month after completion of treatment (post-Tx). The preTx PET/CT and MRI scans were performed at a median of 4 days (range 0–22 days) and 7 days (range 3–22 days) before the start of EBRT, respectively. Mid-Tx PET/CT scans were acquired at a median EBRT dose of 41.4 Gy (range 36–45 Gy), and MRI scans were taken at a median of 1 day (range 0–5 days) before PET/CT. The PET/CT images were acquired 60 min after injection of 5 MBq/kg of FDG. Before PET/CT scans, the patient was asked to fast for at least 6 h and plasma glucose levels were measured in all patients. PET/CT was performed on the Discovery Ste scanner (GE Healthcare, Milwaukee, WI, USA). Non-contrast-enhanced CT images were first acquired using a 16-slice helical CT with AutomA mode (range 30–170 mA), 140 keV, a collimation of 3.75 mm, and a table feed of 17.50 mm/rotation. The scan parameters were 2.5-mm-thick CT images with 2.5-mm spacing. PET emission images of the pelvis were then 3-dimensionally acquired for 2.5 min/frame with an overlap of 9 slices between frames. CT-based attenuation-corrected PET images were reconstructed using a 3D iterative algorithm (VUE Point; GE Healthcare, Milwaukee, WI, USA) and displayed as a 128 9 128 matrix (3.90 9 3.90 mm; slice thickness 3.27 mm). MRI scanning was performed using a 1.5 T (Sigma; GE Healthcare, Milwaukee, WI, USA) or 3

Ann Nucl Med (2013) 27:37–45

T (Intera Achieva 3T; Philips Medical System, Best, The Netherlands) system equipped with a six-channel cardiac sensitivity encoding coil. T1-weighted (axial images), T2weighted (axial, sagittal and coronal images), and dynamic contrast-enhanced T1-weighted sequences (axial and sagittal images) were included. The parameters (1.5T and 3.0T) of T2-weighted images were the following: repetition time/echo time (3300–4575/87–99 ms and 4177–4292/ 90 ms); echo train length (10–12 and 22); slice thickness (5 and 4 mm); slice gap (2 and 0.4 mm). T1-weighted images were obtained using fast or turbo spin echo sequence at the

39

slice profile. The tumor volume before, during, and after CCRT on PET/CT images was expressed as pre-Tx tumor volume (V1PET), mid-Tx tumor volume (V2PET), and postTx residual tumor volume (V3PET), respectively, while the MRI-determined volume was expressed as pre-Tx tumor volume (V1MRI), mid-Tx tumor volume (V2MRI), and postTx residual tumor volume (V3MRI), respectively. The tumor volume reduction rates, expressed as the percent in volume reduction measured on PET/CT and MRI images during and after CCRT relative to the initial tumor volume, were defined in this setting as follows:

Mid-Tx percent volume reduction on PET ¼ ðV1PET  V2PET Þ=V1PET  100 Mid-Tx percent volume reduction on MRI ¼ ðV1MRI  V2MRI Þ=V1MRI  100 Post-Tx percent volume reduction on PET ¼ ðV1PET  V3PET Þ=V1PET  100 Post-Tx percent volume reduction on MRI ¼ ðV1MRI  V3MRI Þ=V1MRI  100:

axial plane. The parameters (1.5T and 3.0T) of T1weighted images were the following: repetition time/echo time (467/12 ms and 437/10 ms); echo train length (2 and 4); slice thickness (5 and 5 mm); slice gap (2 and 2 mm). The parameters (1.5T and 3.0T) of dynamic contrastenhanced image were the following: repetition time/echo time (60/1.4 ms and 8/4.1 ms); flip angle (80° and 25°); slice thickness (5 and 7 mm); slice gap (2 and 0 mm). All image sets from the PET/CT and MRI scans were transferred to a treatment planning system (Pinnacle3, Philips Medical System, The Netherlands). For each patient, three sets of images (pre-Tx PET/CT and MRI, mid-Tx PET/CT and MRI, and post-Tx PET/CT and MRI) were manually co-registered based on pelvic and vertebral bony landmarks using fusion software supplied by the treatment planning system. The GTV was delineated on the PET/CT images using the isocontour of the 40 % threshold of the maximum standardized uptake value (SUVmax). To prevent the inclusion of adjacent intense structures, such as the bladder, lymph nodes, and intestine, the ‘‘region of interest’’ was created manually and the isocontour of the 40 % threshold of SUVmax was drawn. The GTV was also delineated on T2-weighted MRI images. These contours were defined by a radiation oncologist with experience in gynecologic cancer. The lymph nodes were not included in the GTV because, in general, the nodal volumes were too small to be delineated even before treatment. The GTVs of each of the image sets were calculated as the sum of all of the areas of the tumor multiplied by the

Tumor response was defined as follows: complete response (CR), disappearance of tumor; partial response (PR), a decline of at least 65 % in tumor volume; stable disease (SD), neither partial response nor progressive disease; and progressive disease (PD), at least a 70 % increase in tumor volume. The Response Evaluation Criteria in Solid Tumors (RECIST) define PR as a decline of at least 30 % in the sum of all tumor diameters and PD as at least a 20 % increase in tumor diameters [8]. Because a unidimensional measurement of the long axis of tumors is implemented in the RECIST, a 20 % increase in tumor dimensions corresponds to a 72 % increase in the threedimensional volume, and a 30 % decrease in tumor dimensions corresponds to a 65 % decrease in the threedimensional volume. For PET images obtained during treatment, CR was defined as the absence of abnormal FDG uptake at the sites of abnormal FDG uptake in locations consistent with tumor. In MRI images, the tumor response was defined as CR when there was no detectable mass in T2-weighted MRI images. Statistical analysis Statistical analyses were implemented using SPSS software version 18.0. Spearman’s rank correlation was used to compare GTV as determined by PET/CT and MRI. The concordance of responses between MRI and PET/CT of mid-Tx and post-Tx scans was assessed by Cohen’s kappa index. Intra- and inter-image comparisons of the tumor volumes and reduction rates during and after CCRT were

123

40

Ann Nucl Med (2013) 27:37–45

Results

Fig. 1 The initial tumor volume based on MRI versus PET/CT images

Table 2 Treatment response during CCRT (n = 52) PET/CT CR

PR

SD

MRI CR

4

0

PR

10

21

0 2

SD

4

5

6

j = 0.34 ± 0.09 (agreement = 0.60) CCRT concurrent chemoradiotherapy, MRI magnetic resonance imaging, PET positron emission tomography, CT computed tomography, CR complete response, PR partial response, SD stable disease

Table 3 Treatment response after CCRT (n = 52) PET/CT CR

PR

SD

MRI CR

31

2

0

PR

9

7

0

SD

1

2

0

There was a strong correlation between V1PET and V1MRI (correlation coefficient, r2 = 0.814; Fig. 1). The tumor volumes on pre-Tx MRI showed linear correlation and were slightly larger than those on PET/CT (y = 1.155x ? 11.676). During CCRT, PET images showed CR in 18 patients (34.6 %), PR in 26 patients (50.0 %), and SD in 8 patients (15.4 %), while MRI scans showed CR in 4 patients (7.7 %), PR in 33 patients (63.5 %), and SD in 15 patients (28.8 %) (Table 2). After CCRT, 41 patients (78.8 %) achieved CR and 11 patients (21.2 %) achieved PR based on PET/CT images, while 33 patients (63.4 %) achieved CR, 16 patients (30.8 %) achieved PR, and 3 patients (5.8 %) achieved SD based on MRI images (Table 3). Four patients had CR according to both PET/CT and MRI during CCRT, but after completion of CCRT, 31 patients achieved CR according to both PET/CT and MRI images. Figure 2 shows representative examples of MRI and PET/CT scans taken before, during, and after CCRT. The treatment responses evaluated by MRI and PET/CT scans were concordant (Tables 2, 3). The tumor volumes on pre-Tx, mid-Tx, and post-Tx PET/CT scans were significantly different (p \ 0.001). Furthermore, tumor volumes were also significantly different between these three time points based on MRI (p \ 0.001). The geometric mean volume based on MRI scans was 52.0 cm3 pre-Tx and decreased to 7.8 cm3 midTx and 2.0 cm3 post-Tx. In PET/CT images, the geometric mean volume was 36.0 cm3 pre-Tx and decreased to 3.6 cm3 mid-Tx and 1.4 cm3 post-Tx. During CCRT, midTx percent volume reduction on PET was significantly greater than mid-Tx percent volume reduction on MRI (p = 0.024; Fig. 3). However, after completion of CCRT, no significant difference was found in post-Tx percent volume reduction on PET versus post-Tx percent volume reduction on MRI (p = 0.289). The percent volume reduction of adenocarcinomas was significantly smaller than that of squamous cell carcinomas on both PET/CT images (p \ 0.001) and MRI images (p \ 0.001; Fig. 4). Age, FIGO stage, and initial tumor volume did not affect the difference in percent volume reduction of the tumor (Table 4).

Discussion

j = 0.38 ± 0.12 (agreement = 0.73) CCRT concurrent chemoradiotherapy, MRI magnetic resonance imaging, PET positron emission tomography, CT computed tomography, CR complete response, PR partial response, SD stable disease

performed using a repeated measurement ANOVA while controlling for age, stage, initial tumor volume, and pathology. A p value of B0.05 was considered to be statistically significant.

123

Predictors of disease recurrence in cervical cancer include not only clinical stage and lymph node status at the time of initial diagnosis, but also tumor response after treatment [9, 10]. Jacobs et al. [11] first reported in 1986 that the presence of a persistent tumor upon clinical examination 1–3 months after the completion of therapy was an indicator of poor survival outcomes in patients with cervical

Ann Nucl Med (2013) 27:37–45

Fig. 2 Examples of MRI and PET/CT images taken before, during, and after treatment. A 29-year-old female with stage IIB cervical cancer achieved CR on both (a) MRI and PET/CT images (b) during

41

after treatment. Another 48-year-old female with stage IIB cervical cancer achieved PR on c MRI and d CR on PET/CT during treatment and CR on both MRI and PET/CT images after treatment

123

42

Ann Nucl Med (2013) 27:37–45 Table 4 Confounding factors for volume change

Age (years)

Subgroup (n = 52)

Significance for MRI (p value)a

Significance for PET/CT (p value)a

\39 (n = 5)

0.405

0.207

0.427

0.687

0.000 

0.000 

0.322

0.553

40–59 (n = 30) C60 (n = 17) FIGO stage

IB1–IIA (n = 8) IIB–IVA (n = 44)

Pathology

SQ (n = 47) AD (n = 5)

Fig. 3 Relative volume change according to MRI and PET/CT with 95 % confidence interval during and after treatment (Tx)

Initial tumor volume (cm3)

\52 (n = 24) C52 (n = 28)

MRI magnetic resonance imaging, PET positron emission tomography, CT computed tomography, SQ squamous cell carcinoma, AD adenocarcinoma

Fig. 4 Relative volume change based on MRI and PET/CT images of squamous cell carcinomas (SQ) and adenocarcinomas (AD) with 95 % confidence interval during and after treatment (Tx)

cancer. The 5-year survival outcome of patients with no appreciable tumor was 76 %, compared to 42 % for those with findings suggestive of a tumor and 8 % for those with a persistent tumor (p \ 0.001). In a prospective cohort study, Schwarz et al. [4] demonstrated that 3-month posttreatment FDG uptake was predictive of survival. In this study, patients were imaged with PET between 2 and 4 months (mean, 3 months) after the completion of treatment for cervical cancer. The 3-year progression-free survival rates of metabolic CR, PR, and PD patients were 78, 33, and 0 %, respectively (p \ 0.001).

123

 

p \ 0.0001

a

Repeated measures ANOVA

However, the prognostic information obtained after the completion of treatment has the least practical impact in clinical practice, because the opportunity for treatment modification has passed. If tumor response could be predicted or assessed during or just after the course of therapy, treatment could be adapted and individualized. Patients with persistent and large tumors during the course of therapy could be identified earlier, and chemotherapy regimens could be altered accordingly. Conversely, patients with no evidence of residual PET activity during the course of their therapy may be candidates for dose alterations that could potentially result in reduced acute and late toxicity. Assessment of tumor size by MRI during treatment has been shown by several groups to be an early predictor of outcome in patients with cervical cancer [6, 12]. In a study by Mayr et al. [12], the rate of tumor regression was found to correlate with treatment outcome. In their study, images were collected before treatment initiation and after 4–5 weeks of RT at 40–50 Gy, which corresponded to immediately prior to the delivery of brachytherapy in their treatment schema. Patients whose tumors regressed to less than 20 % of the initial volume had a cumulative incidence of local recurrence of 9.5 %, as compared to 76.9 % in patients whose tumors regressed to more than 20 % of the residual volume (p \ 0.001). We have also previously

Ann Nucl Med (2013) 27:37–45

reported that the tumor volume regression rate during RT as measured by MRI was a significant predictor of local control rate in cervical cancer patients [6]. There are few studies about the tumor volume discrepancies between MRI and PET/CT in cervical cancer patients at the time of diagnosis. Ma et al. [7] reported that the MRI and FDG-PET tumor volumes were similar, whereas the location of the tumor (volume) varied—especially for small tumor volumes. Figure 1 shows strong correlation between the initial tumor volume of PET and that of MRI. The volume regression rate during RT by MRI was a significant predictor of local control rate in cervical cancer patients. So, we hypothesized that the tumor volume as measured by PET/CT could also be a possible predictor of local control rate in cervical cancer patients. Schwarz et al. [13] proposed the feasibility of FDG-PET as a monitor of treatment response for cervical cancer during radiation therapy. Several studies have used PET to assess tumor response during RT of other malignancies. Kong et al. [14] reported a significant reduction of tumor FDG activity in patients with lung cancer undergoing RT without a significant difference in FDG activity within the irradiated lung during RT. They observed that the metabolic response and the peak FDG activity of the tumor during RT at approximately 45 Gy were highly correlated with the ultimate responses 3–4 months after completion of RT. Wieder et al. [15] reported that during preoperative CCRT of esophageal cancer, changes in FDG uptake by a tumor after 14 days of therapy could predict the subsequent histopathologic tumor response and were significantly correlated with overall survival. They also reported that there was a 59 % decrease in baseline metabolic activity immediately after completion of CCRT, with no additional decrease in FDG uptake afterward. We found that the percent volume reduction calculated from PET/CT images was significantly greater than that calculated from MRI images during CCRT, but there was no significant difference in the percent volume reduction on PET/CT versus MRI after completion of CCRT. However, the timing of mid-Tx MRI scans in our study, namely 0–5 days before PET/CT scans, might have resulted in the differences in tumor volume measured by PET/CT and MRI during CCRT for cervical cancer. The optimal timing to obtain PET scans during RT for cervical cancer is unclear. FDG activity tends to show an initial increase followed by a steady decline after treatment [16]. Early elevation of FDG activity may be associated with acute inflammation in normal tissue or elevated metabolic activity within the tumor cells; the later reduction in FDG uptake is a reflection of a decrease in the number of viable tumor cells or a reduction in tumor metabolic activity. In our institution, brachytherapy

43

simulations by PET/CT are routinely used as a part of standard care. Mid-Tx PET/CTs were acquired at a median EBRT dose of 41.4 Gy (range 36–45 Gy) for brachytherapy simulation, and mid-Tx MRI images were taken on the same day or a few days (median 1 day) before brachytherapy simulation. Lin et al. [5] reported that a 50 % physiologic tumor volume reduction occurred within 20 days of the initiation of therapy in cervical cancer patients. Hatano et al. [17] reported that patients with a tumor size reduction under 30 % by MRI-based volume measurements at 30 Gy EBRT gained local control. Mayr et al. [12] reported that the tumor regression rate by MRI at 40–50 Gy was significantly correlated with the 5-year local control and the disease-free survival rate. According to these studies, our timing of mid-Tx PET/CT and MRI scans during the fourth week of EBRT is reasonable. However, there is no consensus relating to the time points at which PET should be performed to achieve the most accurate assessment of tumor response. Recommendations regarding the optimal timing of PET during treatment for cervical cancer will require more systematic study. Guidelines for the appropriate delineation of GTV on the PET/CT image are still being developed. Commonly used methods for determining tumor volume using PET are the qualitative visual method, the fixed SUV value, the anatomic biologic contouring method, and the GTV 40 % of SUVmax method [18]. In our study, we chose the 40 % isointensity level. For cervical cancer, Miller et al. [19] found that PET-GTV and CT-GTV were approximately equal using a 40 % threshold of the SUVmax. The absolute tumor volume as measured by MRI and PET/CT may differ because of the different natures of the two imaging modalities. Unlike anatomic imaging modalities such as CT or MRI that provide the anatomic extent of morphologic changes within a patient’s body, PET provides both a spatial and quantitative representation of specific metabolic activity, providing a basis for quantitative functional imaging analysis. Paulino et al. [20] reported that PET-GTV was smaller than CT-GTV in most head and neck cancer cases that they examined. In our study, the tumor volumes calculated from pretreatment PET/CT images were slightly smaller than those calculated from MRI images. In our study, 5 of 52 patients had adenocarcinoma, and the tumor reduction rate was significantly smaller in these patients than in the squamous cell carcinoma patients. The number of patients in the current study with adenocarcinoma was too small for us to generalize these findings, but it may be that the poor volume reduction of adenocarcinomas is correlated to the poor prognosis associated with this particular tumor pathology. It is controversial as to whether histologic type is an independent prognostic factor for survival. However, the

123

44

majority of studies have shown that patients with adenocarcinomas have a worse prognosis than those with squamous cell carcinomas with a 10–20 % difference in the 5-year overall survival rate [21]. Tumor response evaluation in our study differed from that based on the WHO or RECIST, in that the image data were treated as three-dimensional volumetric processes. This study was designed as a prospective trial and had a homogeneous study population and a uniform treatment regimen. Although the follow-up interval in this study was too short to be able to perform local control and survival analyses, the tumor volume change between mid-Tx and post-Tx was evident much earlier on PET/CT scans than on MRI scans. The tumor volume regression rate during RT measured by MRI was a significant predictor of local control rate [6]; therefore, the tumor volume regression rate during RT measured by PET/CT may be a significant predictor of the local control rate in uterine cervical cancer patients. There were a few studies on early PET/CT response during treatment in cervical cancer patients [22, 23]. None of the authors could show any definite evidence of the usefulness of early PET response for local control or survival yet. But these data showed some important tips for the evaluation of early PET response. Bjurberg et al. [22] showed that early metabolic response at 11–24 days could identify only a small group (22 %, 7 of 32 patients) with excellent prognosis. They suggested that a later time point for cervical cancer might be chosen for the clinical value. Kidd et al. [23] have recently suggested that week 4 of treatment could be the best time point for the prediction of response. They showed that week 4 SUVmax was significantly associated with post-treatment PET response at 3 months (p = 0.037). Although we could not perform a survival analysis in this study because of short follow-up time, there was a possibility that good response at mid-Tx would show a favorable treatment outcome. Further studies are needed to determine whether changes in tumor volume during and immediately after CCRT on PET/CT and MRI scans are associated with the local control rate and survival outcome. It is unclear whether PET and MRI can substitute for one another or complement one another to predict the treatment response and prognosis after CCRT in cervical cancer patients.

Ann Nucl Med (2013) 27:37–45

2.

3. 4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

Conflict of interest The authors have no conflicts of interest in connection with this work. 16. 17.

References 1. Hricak H, Gatsonis C, Coakley FV, Snyder B, Reinhold C, Schwartz LH, et al. Early invasive cervical cancer: CT and MR imaging in preoperative evaluation—ACRIN/GOG comparative

123

18.

study of diagnostic performance and interobserver variability. Radiology. 2007;245:491–8. Showalter TN, Miller TR, Huettner P, Rader J, Grigsby PW. 18Ffluorodeoxyglucose-positron emission tomography and pathologic tumor size in early-stage invasive cervical cancer. Int J Gynecol Cancer. 2009;19:1412–4. Grigsby PW. The role of FDG-PET/CT imaging after radiation therapy. Gynecol Oncol. 2007;107:S27–9. 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. Lin LL, Yang Z, Mutic S, Miller TR, Grigsby PW. FDG-PET imaging for the assessment of physiologic volume response during radiotherapy in cervix cancer. Int J Radiat Oncol Biol Phys. 2006;65:177–81. Nam H, Park W, Huh SJ, Bae DS, Kim BG, Lee JH, et al. The prognostic significance of tumor volume regression during radiotherapy and concurrent chemoradiotherapy for cervical cancer using MRI. Gynecol Oncol. 2007;107:320–5. Ma DJ, Zhu JM, Grigsby PW. Tumor volume discrepancies between FDG-PET and MRI for cervical cancer. Radiother Oncol. 2011;98:139–42. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst. 2000;92:205–16. Hardt N, van Nagell JR, Hanson M, Donaldson E, Yoneda J, Maruyama Y. Radiation-induced tumor regression as a prognostic factor in patients with invasive cervical cancer. Cancer. 1982;49:35–9. Hong JH, Chen MS, Lin FJ, Tang SG. Prognostic assessment of tumor regression after external irradiation for cervical cancer. Int J Radiat Oncol Biol Phys. 1992;22:913–7. Jacobs AJ, Faris C, Perez CA, Kao MS, Galakatos A, Camel HM. Short-term persistence of carcinoma of the uterine cervix after radiation. An indicator of long-term prognosis. Cancer. 1986;57:944–50. Mayr NA, Magnotta VA, Ehrhardt JC, Wheeler JA, Sorosky JI, Wen BC, et al. Usefulness of tumor volumetry by magnetic resonance imaging in assessing response to radiation therapy in carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys. 1996;35:915–24. Schwarz JK, Lin LL, Siegel BA, Miller TR, Grigsby PW. 18-Ffluorodeoxyglucose-positron emission tomography evaluation of early metabolic response during radiation therapy for cervical cancer. Int J Radiat Oncol Biol Phys. 2008;72:1502–7. Kong FM, Frey KA, Quint LE, Ten Haken RK, Hayman JA, Kessler M, et al. A pilot study of [18F]fluorodeoxyglucose positron emission tomography scans during and after radiationbased therapy in patients with non-small-cell lung cancer. J Clin Oncol. 2007;36:3116–23. Wieder HA, Bru¨cher BL, Zimmermann F, Becker K, Lordick F, Beer A, et al. Time course of tumor metabolic activity during chemoradiotherapy of esophageal squamous cell carcinoma and response to treatment. J Clin Oncol. 2004;22:900–8. Hautzel H, Mu¨ller-Ga¨rtner HW. Early changes in fluorine-18FDG uptake during radiotherapy. J Nucl Med. 1997;38:1384–6. Hatano K, Sekiya Y, Araki H, Sakai M, Togawa T, Narita Y, et al. Evaluation of the therapeutic effect of radiotherapy on cervical cancer using magnetic resonance imaging. Int J Radiat Oncol Biol Phys. 1999;45:639–44. Devic S, Tomic N, Faria S, Menard S, Lisbona R, Lehnert S. Defining radiotherapy target volumes using 18F-fluoro-deoxy-

Ann Nucl Med (2013) 27:37–45 glucose positron emission tomography/computed tomography: still a Pandora’s box? Int J Radiat Oncol Biol Phys. 2010;78: 1555–62. 19. Miller T, Grigsby P. Measurement of tumor volume by PET to evaluate prognosis in patients with advanced cervical cancer treated with radiation therapy. Int J Radiat Oncol Biol Phys. 2002;53:353–9. 20. Paulino AC, Koshy M, Howell R, Schuster D, Davis LW. Comparison of CT- and FDG-PET-defined gross tumor volume in intensity-modulated radiotherapy for head-and-neck cancer. Int J Radiat Oncol Biol Phys. 2005;61:1385–92.

45 21. Gien LT, Beauchemin MC, Thomas G. Adenocarcinoma: a unique cervical cancer. Gynecol Oncol. 2010;116:140–6. 22. Bjurberg M, Kjelle´n E, Ohlsson T, Bendahl PO, Brun E. Prediction of patient outcome with 2-deoxy-2-[18F]fluoro-D-glucosepositron emission tomography early during radiotherapy for locally advanced cervical cancer. Int J Gynecol Cancer. 2009;19:1600–5. 23. Kidd EA, Thomas M, Siegel BA, Dehdashti F, Grigsby PW. Changes in cervical cancer FDG uptake during chemoradiation and association with response. Int J Radiat Oncol Biol Phys. 2012 (Epub ahead of print).

123