Diagnosis performance of positron emission tomography-computed tomography among cervical cancer patients

Abstract

OBJECTIVE: This study aims to investigate the value of PET/CT scanning in preoperative diagnosis of cervical carcinoma, especially for detecting lymph node metastasis.

MATERIALS AND METHODS: The approval of the ethics committee which was granted by our hospital was obtained beforehand. Patients with CC were collected into this investigation between January 2011 and October 2015. Each participant received surgeries, as well as pelvic and paraaortic systematic lymph node dissection. After operations, CC types were confirmed by pathological examinations. The tumor stages were assessed by 3 experienced radiologists independently, according to FIGO examinations and positron emission tomography-computed tomography (PET/CT), and these above diagnostic results were compared with postoperative biopsy pathology, respectively. Statistical analysis was done by SPSS 16.0, and the diagnostic performance of PET/CT was calculated.

RESULTS: 51 patients were identified in this investigation, and the mean age of these female individuals was 42.3±6.7 years (range, 34–58 years). Depending on statistical analysis, the staging accuracy of PET/CT to detect primary tumors was 84.31%, with sensitivity 88.00% and specificity 80.77%. With respect to lymph nodes, the accuracy could reach 76.47%, with sensitivity 82.61% and specificity 71.43%. On the other hand, FIGO staging performed poorly in detecting primary tumors, with sensitivity 44.12%, specificity 47.06% and accuracy 45.10%. In terms of testing lymph nodes, the diagnosis parameters were as followed, sensitivity 28.57%, specificity 8.70% and accuracy 19.61%. After statistical analysis, there was significantly different between two methods (P < 0.05).

CONCLUSIONS: PET/CT scanning may be valuable in detecting primary tumors and lymph nodes, and more accurate staging may lead to improving therapeutic planning in CC patients.

Keywords

Diagnosis positron emission tomography-computed tomography cervical cancer

1 Introduction

As we all know, cervical cancer (CC) is the third most commonly diagnosed carcinoma, and is the most common gynecologic malignancy among female individuals around the world [1], and represent an important cause of morbidity and death among females. In 2014, the American Cancer Society issued the statistical data, 12360 new cases and 4020 deaths for invasive CC occurred in the United States [2]. Nowadays, how to improve the overall survival for patients suffering from CC is always the most disturbing problem for gynecologists and oncologists. It is acknowledged that the choice of treatment mainly depends on the staging when patients were initially diagnosed, and the precise understanding of tumor extent is indeed vital for therapeutic decision-making [3, 4].

With respect to diagnosis, in general, CC is clinically staged by the International Federation of Gynaecology and Obstetrics (FIGO) classification system [5], including chest radiography, transvaginal ultrasound, clinical and gynaecological examinations, etc. Yet, FIGO staging is not good at distinguishing lymph node metastasis (LNM), which is an important factor in treatment planning and prediction of prognosis for CC patients [6]. Currently, several studies have illustrated that PET-CT (positron emission tomography-computed tomography) takes the advantage of higher sensitivity in detecting LNM, when compared with other currently used imaging techniques [7 –11]. And PET-CT has also been proposed for inclusion in the current clinical practice guidelines by gynecological oncologists, mainly to detect distant metastases [12]. Additionally, it is reported that the PET/CT technology may be contribute to evaluate disease-free survival (DFS), time-to-relapse (TTR), and overall survival (OS) [13].

However, the diagnostic accuracy of PET/CT for LNM varied among studies [14]. Moreover, population heterogeneity and other potential factors might also influence the performance of PET/CT for CC staging. To explore whether PET/CT is of particular value in detecting LNM, a retrospective investigation were performed in our department, on the basis of the final pathological diagnosis.

2 Methods

2.1 Study design

We conducted a retrospective study and included patients diagnosed with CC in the period between January 2011 and October 2015. Inclusion criteria were as followed: (1) new diagnosis of CC patients; (2) stage IB-VIA on the basis of FIGO classification system; (3) all patients had PET-CT imaging at the time of diagnosis; (4) all the patients received pelvic and paraaortic systematic lymph node dissection; (5) postoperative pathological examination to establish definitive diagnoses; (6) age 18 years or older. Exclusion criteria were as followed: (1) patients who were pregnant before diagnosis or during the follow-up period; (2) another simultaneous malignancy was diagnosed, such as mental illness; (3) the patients who were included into other clinical trials. Because this research was a retrospective study based on previous clinical data, informed consent could not be obtained from each patient. Nevertheless, the approval of the ethics committee which was granted by our hospital was obtained beforehand. Patient information was collected and evaluated retrospectively. And, the data extraction included: age at diagnosis, date of PET/CT, clinical FIGO stage, histological type and grade, presence and site of positive pelvic lymph nodes, adjuvant therapy (radiotherapy, chemotherapy, physical therapy, etc.).

2.2 PET/CT protocol and pathological evaluations

A retrospective and comprehensive review was carried out by three experienced radiologists from our hospital, and they were all blinded to the pathological results. The relevant parameters of PET/CT were as followed: (1) The instrument is 16-section Discovery STE PET-CT system (GE Healthcare, Amersham, UK), whose brand version is Biograph CT system (Siemens Medical Solutions, Knoxville, TN, USA), and then data were transferred to an AZE workstation (AZE Ltd, Tokyo, Japan). (2) PET scanning was performed at 60 minutes following intravenous injection of 450 MBq dose of FDG. (3) CT scanning was carried out in the spiral mode with 80 mAs current, 140 kV voltage, 0.85 mm collimation, and 15 mm per gantry rotation. (4) No iodinated contrast material was given to participants. After surgeries, dissected lymph nodes were fixed in 10% formalin and then stained with hematoxylin and eosin to test the LNM. The pathological diagnoses were made by 3 qualified pathologists and they were blinded to the PET/CT results.

2.3 Statistical analysis

The extracted data were analyzed using SPSS 16.0 software (ver. 21, New York, NY, USA). And, the diagnostic accuracies of PET/CT for detecting LNM were calculated for all CC cases. 95% confidence intervals (CIs) were generated using the Clopper-Pearson method, P values < 0.05 was considered to be statistically significant. The diagnostic performance which includes sensitivity, specificity, and positive/negative predictive values, was calculated. And, these above indexes were estimated using the following formulas. Sensitivity = true positive/ (true positive + false negative); Specificity = true negative/ (true negative + false positive); Positive Predictive Value = true positive/ (true positive + false positive); Negative Predictive Value = true negative/ (true negative + false negative); Accuracy = (true positive + true negative)/ total number.

3 Results

3.1 Patients

Eighty-five CC patients underwent PET/CT scanning for the initial staging, and thirty-four individuals were excluded because they did not have preoperative information to analysis. The characteristics of the patients are displayed in Table 1. In detail, the mean age was 42.3±6.7 years, the most common pathological type was squamous cell carcinoma (N = 29, 56.86%). The FIGO staging was as followed: IB (N = 11), IIA (N = 1), IIB (N = 10), IIIA (N = 7), IIIB (N = 7), VIA (N = 15). According patient records, there were no adverse events associated with PET/CT examinations.

3.2 Diagnosis accuracy of PET/CT and FIGO classification system

Final pathologic analysis revealed the diagnostic performance of PET/CT, and then compared with that of FIGO classification system. As shown in Table 2, concerning primary tumors, the accuracy of preoperative FIGO clinical examination is 59.15%, and 19.61% for detecting LN. On the contrast, the PET/CT technology performed better, and the detailed analysis were as followed. As shown in Table 3, in terms of detecting primary tumors, in 43 (84.31%) of the 51 cases, PET/CT and pathological classified identically, in 28 (54.90%) cases PET/CT could not stage accurately. After statistical analysis, the parameters showed a significant difference (PET/CT vs. FIGO staging: sensitivity P = 0.026; specificity P = 0.045; accuracy P < 0.01). Furthermore, it is worth mentioning that PET/CT was superior to conventional FIGO staging in testing LNM. In 39 (76.47%) of the 51 cases, PET/CT and pathological classified identically, in 12 (23.53%) cases PET/CT could not stage accurately. Statistically significant was found after statistical analysis (PET/CT vs. FIGO staging: sensitivity P < 0.0001; specificity P < 0.0001; accuracy P < 0.0001).

4 Discussions

Through this retrospective analysis, we compared the diagnosis capability of PET/CT and FIGO classification system, setting biopsy examinations as the golden standard. According to the statistical analysis, we found PET/CT technology appeared superior, with a higher overall accuracy and sensitivity in the detecting tumors, when compared with the conventional staging methods. Similar to other previous studies [15, 16], PET/CT probably plays an important role for diagnosing CC, not only in improving diagnosis but also in optimizing the choice of treatment.

Nevertheless, there are some notable limitations in this study, and should be proposed. Firstly, the largest limitation is the relatively small sample size and its retrospective nature, together with the period lasting nearly 5 years, all of which might weaken the power of test to some extent. Secondly, as far as I am concerned, the results could not reflect the actual diagnosis performance of PET/CT, and the efficacy may be overestimated sometimes, because PET evaluates the metabolic features of cancer cells and the diagnosis performance may be influenced by tissue types and pathological types, etc. [17]. Thirdly, another challenging problem is the limited accuracy of PET/CT in terms of detecting smaller lymph nodes. For example, Gouy S et al. [18] conducted a multicenter prospective study, and 237 patients with PET/CT negative para-aortic lymph nodes received a laparoscopic lymphadenectomy. After pathological correlations, the authors found that the PET/CT was falsely negative in 12% of patients, especially for lymph nodes with ≤5 mm diameter [19]. Besides, not every patient received PET/CT scanning at our hospital, but the majority of them were. Thus, the inherent heterogeneity of the diagnosis results could not be avoided. Moreover, the lymph node region can also strongly affect the efficacy of PET/CT to some extent [6]. At last but not the least, there is growing interests in PET and MRI fusion imaging to evaluate and follow-up CC patients more effectively, and we hope other organizations to become involved in similar endeavors.

5 Conclusions

From the data reported above, the conclusion may be highlighted that the diagnostic ability of PET/CT in detection of primary tumors and lymph nodes is superior with higher sensitivity and specificity to that of traditional FIGO classification system, especially for LNM. Furthermore, PET/CT should be included in the diagnostic guideline to contribute for more appropriate therapeutic planning for CC patients.

Conflict of interest statement

The authors declare no conflicts of interest.

Footnotes

Acknowledgments

This study received financial support from Shenzhen Science and Technology Project (project number JCYJ20140414170821311 and JCYJ20140414170821323).

References

Baker

, Pelkofski

and Te Paske

, et al., Preoperative imaging of uterine malignancy: A low-value service, Gynecol Oncol 137(7) (2015), 285–290.

Musto

, Grassetto

and Marzola

M.C.

, et al., Role of 18F-FDG PET/CT in the carcinoma of the uterus: A review of literature, Yonsei Med J 55(7) (2014), 1467–1472.

Rauch

G.M.

, Kaur

and Choi

, et al., Optimization of MR imaging for pretreatment evaluation of patients with endometrial and cervical cancer, Radiographics 34(7) (2014), 1082–1098.

Ferrandina

, Petrillo

and Restaino

, et al., Can radicality of surgery be safely modulated on the basis of MRI and PET/CT imaging in locally advanced cervical cancer patients administered preoperative treatment? Cancer 118(2) (2012), 392–403.

Pecorelli

, Revised FIGO staging for carcinoma of the vulva, cervix, and endometrium, Int J Gynaecol Obstet 105(2) (2009), 103–104.

Nogami

, Banno

and Irie

, et al., The efficacy of preoperative positron emission tomography-computed tomography (PET-CT) for detection of lymph node metastasis in cervical and endometrial cancer: Clinical and pathological factors influencing it, Jpn J Clin Oncol 45(7) (2015), 26–34.

Lazzari

, Cecconi

and Jereczek-Fossa

B.A.

, et al., The role of [(18)F]FDG-PET/CT in staging and treatment planning for volumetric modulated Rapidarc radiotherapy in cervical cancer: Experience of the European Institute of Oncology, Milan, Italy, Ecancermedicalscience 8 (2014), 405.

Kitajima

, Suzuki

and Senda

, et al., Preoperative nodal staging of uterine cancer: Is contrast-enhanced PET/CT more accurate than non-enhanced PET/CT or enhanced CT alone? Ann Nucl Med 25(7) (2011), 511–519.

Chou

H.H.

, Chang

H.P.

and Lai

C.H.

, et al., (18)F-FDG PET in stage IB/IIB cervical adenocarcinoma/adenosquamous carcinoma, Eur J Nucl Med Mol Imaging 37(7) (2010), 728–735.

10.

Choi

H.J.

, Ju

and Myung

S.K.

, et al., Diagnostic performance of computer tomography, magnetic resonance imaging, and positron emission tomography or positron emission tomography/computer tomography for detection of metastatic lymph nodes in patients with cervical cancer: Meta-analysis, Cancer Sci 101(7) (2010), 1471–1479.

11.

Vandeperre

, Van Limbergen

and Leunen

, et al., Para-aortic lymph node metastases in locally advanced cervical cancer: Comparison between surgical staging and imaging, Gynecol Oncol 138(7) (2015), 299–303.

12.

Colombo

, Carinelli

and Colombo

, et al., Cervical cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up, Ann Oncol 23(Suppl 7) (2012), vii27–vii32.

13.

Kim

B.S.

, Kim

I.J.

and Kim

S.J.

, et al., The prognostic value of the metabolic tumor volume in FIGO stage IA to IIB cervical cancer for tumor recurrence: Measured by F-18 FDG PET/CT, Nucl Med Mol Imaging 45(7) (2011), 36–42.

14.

Nogami

, Iida

and Banno

, et al., Application of FDG-PET in cervical cancer and endometrial cancer: Utility and future prospects, Anticancer Res 34(7) (2014), 585–592.

15.

Roh

J.L.

, Kim

J.S.

and Lee

J.H.

, et al., Utility of combined (18)F-fluorodeoxyglucose-positron emission tomography and computed tomography in patients with cervical metastases from unknown primary tumors, Oral Oncol 45(7) (2009), 218–224.

16.

Rusthoven

K.E.

, Koshy

and Paulino

A.C.

, The role of fluorodeoxyglucose positron emission tomography in cervical lymph node metastases from an unknown primary tumor, Cancer 101(11) (2004), 2641–2649.

17.

Kidd

E.A.

, Spencer

C.R.

and Huettner

P.C.

, et al., Cervical cancer histology and tumor differentiation affect 18F-fluorodeoxyglucose uptake, Cancer 115(7) (2009), 3548–3554.

18.

Gouy

, Morice

and Narducci

, et al., Prospective multicenter study evaluating the survival of patients with locally advanced cervical cancer undergoing laparoscopic para-aortic lymphadenectomy before chemoradiotherapy in the era of positron emission tomography imaging, J Clin Oncol 31(7) (2013), 3026–3033.

19.

Sironi

, Buda

and Picchio

, et al., Lymph node metastasis in patients with clinical early-stage cervical cancer: Detection with integrated FDG PET/CT, Radiology 238(7) (2006), 272–279.