The detailed characteristics of the enrolled patients are shown in Table 1. According to our inclusion and exclusion criteria, 125 of the 1560 patients were finally included in this study. A total of 112 patients (89.6%) presented with stage IIB or above cervical cancer. A total of 114 patients (91.2%) had squamous cell carcinoma, 9 patients (7.2%) had adenocarcinoma, 1 patient had clear cell carcinoma, and the remaining patient had Mullerian carcinosarcoma. Fifty-two patients (41.6%) had a tumour size greater than 4 cm by gynaecological examination. Forty-three patients (34.4%) had positive pelvic lymph nodes (PLNs) and 2 patients (1.6%) had positive para-aortic lymph nodes (PALNs) confirmed by 18F-FDG PET/CT; 10 patients (8%) with positive PALNs had concomitant PLN metastasis.
Table 1
The characteristics of the enrolled patients
Characteristics |
Total number of patients | 125 |
Age[median(range)](years) | 50(30–81) |
FIGO(2009) stage | |
IB | 10 (8%) |
IIA | 3 (2.4% ) |
IIB | 84 (67.2%) |
IIIA | 3 (2.4%) |
IIIB | 24 (19.2%) |
IVA | 1 (0.8%) |
Histology | |
Squamous cell carcinoma | 114 (91.2%) |
Adenocarcinoma | 9 (7.2%) |
others | 2 (1.6%) |
Tumour size(cm) | |
≤ 4 | 73 (58.4%) |
༞4 | 52 (41.6%) |
Pelvic LN involvement | 43 (34.4%) |
Para-aortic LN involvement | 2 (1.6%) |
Both pelvic LN and para-aortic LN involvement | 10 (8%) |
Overall treatment time(days) | |
≤ 56 | 96 (76.8%) |
༞56 | 29 (23.2%) |
Total dose of brachytherapy(Gy) | |
< 30(fraction range of 5-7Gy) | 14 (11.2%) |
≥ 30(fraction range of 5-7Gy) | 111 (88.8%) |
Number of concurrent chemotherapy(fractions) | |
< 4 | 28 (22.4%) |
≥ 4 | 97 (77.6%) |
SUVmean | |
< 7.9 | 74 (59.2%) |
≥ 7.9 | 51 (40.8%) |
SUVmax | |
< 12.8 | 68 (54.4%) |
≥ 12.8 | 57 (45.6%) |
MTV(cm3) | |
< 18.3 | 59 (47.2%) |
≥ 18.3 | 66 (52.8%) |
TLG | |
< 113.4 | 55 (44%) |
≥ 113.4 | 70 (56%) |
Abbreviations: FIGO = International Federation of Gynecology and Obstetrics; LN = lymph node; SUVmean = mean standardized uptake value; SUVmax = maximum standardized uptake; MTV = metabolic tumor volume; value; TLG = total lesion glycolysis. |
All 125 patients completed radiotherapy with a median time of 51 days (range, 42–98 days). Twelve patients (9.6%) received neoadjuvant chemotherapy followed by concurrent chemoradiotherapy or radiotherapy alone, 102 patients (81.6%) received concurrent chemoradiotherapy as the primary therapy, and the remaining 11 patients (8.8%) received radiotherapy alone. Ninety-seven patients (77.6%) completed more than or equal to four cycles of chemotherapy. A total of 111 patients (88.8%) underwent a total dose of brachytherapy greater than or equal to 30 Gy at a single dose range of 5–7 Gy.
The median follow-up period was 62 months (range, 4-114 months). A total of 30 patients (24%) suffered from treatment failure, including 6 patients with local recurrence, 21 patients with distant failure, and 3 patients with concurrent local and distant failure. The total local recurrence and distant failure rates were 7.2% and 19.2%, respectively. The most common site of local recurrence was the cervix uterus, and the most common site of distant metastasis was the lung. The 5-year OS, DFS, LC and DMFS rates were 83.6%, 75.1%, 92.3% and 79.9%, respectively (Fig. 1).
ROC curve analysis was used to determine the best cut-off values of SUVmax, SUVmean, MTV, and TLG in predicting the prognosis of cervical cancer, considering the sensitivity and specificity for DFS. The areas under the curves of SUVmax, SUVmean, MTV, and TLG were 0.53 (p = 0.595; 95% CI 0.41–0.65), 0.54 (p = 0.544; 95% CI 0.42–0.66), 0.57 (p = 0.267; 95% CI 0.45–0.68), and 0.57 (p = 0.223; 95% CI 0.46–0.69), respectively. The optimal cut-off points of SUVmax, SUVmean, MTV and TLG were 12.8, 7.9, 18.3 cm3 and 113.4, respectively.
The univariate analysis showed that PALN, total dose of brachytherapy, and TLG were significantly associated with DFS (Table 2). After multivariate analysis, PALN (HR 0.12; 95% CI, 0.05–0.28; p < 0.001), total dose of brachytherapy ≥ 30 Gy (HR 3.30; 95% CI, 1.32–8.25; p = 0.011), and TLG level ≥ 113.4 (HR 0.28; 95% CI, 0.12–0.64; p = 0.003) remained significant in predicting DFS (Table 3). FIGO stage, PALN, total dose of brachytherapy, and TLG were independent prognostic factors for OS in multivariate analysis. PALN was a poor prognostic factor for LC. PALN, total dose of brachytherapy, and TLG had important impacts on DMFS in multivariate analysis. MTV was an important prognostic factor for OS and DMFS in univariate analyses; however, no significant differences were identified for OS and DMFS in multivariate analyses. Moreover, SUVmax and SUVmean had no significant influence on DFS, OS, LC, or DMFS in either univariate or multivariate analysis.
Table 2
Univariate analyses of clinical factors and PET metabolic parameters for disease-free survival
variable | Univariate analysis |
| HR | 95%CI | p |
Age(continuous, year) | 1.003 | 0.969–1.038 | 0.882 |
FIGO(2009) stage | | | |
Ⅰ-Ⅱ vs. Ⅲ-Ⅳ | 2.114 | 0.988–4.523 | 0.054 |
Histology | | | |
Squamous vs. non-squamous | 2.459 | 0.939–6.440 | 0.067 |
Tumour size(cm) | | | |
≤ 4 vs. ༞4 | 1.336 | 0.651–2.738 | 0.429 |
Pelvic LN involvement | | | |
Negative vs. Positive | 1.368 | 0.667–2.805 | 0.393 |
Para-aortic LN involvement | | | |
Negative vs. Positive | 6.166 | 2.711–14.024 | < 0.001 |
Overall treatment time(days) | | | |
≤ 56 vs. ༞56 | 1.785 | 0.835–3.818 | 0.135 |
Total dose of brachytherapy(Gy) | | | |
< 30 vs. ≥30(fraction range of 5-7Gy) | 0.382 | 0.155–0.942 | 0.037 |
Number of concurrent chemotherapy(fractions) | | | |
< 4 vs. ≥4 | 0.839 | 0.360–1.956 | 0.684 |
SUVmean | | | |
< 7.9 vs. ≥7.9 | 1.555 | 0.759–3.187 | 0.227 |
SUVmax | | | |
< 12.8 vs. ≥12.8 | 1.788 | 0.867–3.688 | 0.116 |
MTV(cm3) | | | |
< 18.3 vs. ≥18.3 | 2.095 | 0.980–4.480 | 0.056 |
TLG | | | |
< 113.4 vs. ≥113.4 | 2.629 | 1.169–5.914 | 0.019 |
Abbreviations: HR = hazard ratio; CI = confidence interval; FIGO = International Federation of Gynecology and Obstetrics; LN = lymph node; SUVmean = mean standardized uptake value; SUVmax = maximum standardized uptake; MTV = metabolic tumor volume; TLG = total lesion glycolysis. |
Table 3
multivariate analyses of clinical factors and PET metabolic parameters for disease-free survival
variable | multivariate analysis |
| HR | 95%CI | p |
Para-aortic LN involvement | | | |
Negative vs. Positive | 0.116 | 0.048–0.278 | < 0.001 |
Total dose of brachytherapy(Gy) | | | |
< 30 vs. ≥ 30(fraction range of 5-7Gy) | 3.296 | 1.316–8.253 | 0.011 |
TLG | | | |
< 113.4 vs. ≥ 113.4 | 0.278 | 0.121–0.640 | 0.003 |
Abbreviations: HR = hazard ratio; CI = confidence interval; LN = lymph node; TLG = total lesion glycolysis. |
The 5-year OS, DFS, LC and DMFS rates for patients with TLG levels < 113.4 and ≥ 113.4 were 90.1% and 78% (p = 0.055, Fig. 2), 86.5% and 66.1% (p = 0.015, Fig. 3), 96.2% and 89.3% (p = 0.136, Fig. 4), and 90.9% and 71.2% (p = 0.02, Fig. 5), respectively. For the 70 patients with TLG ≥ 113.4, the median follow-up period was 61 months (range, 4-110 months). The median DFS period was 56 months. Of these patients, 22 patients (31.4%) experienced treatment failure, including 4 patients with local failure, 15 patients with distant failure, and 3 patients with concurrent local and distant failure. The total local failure and distant failure rates were 10% and 25.7%, respectively. Of the 22 patients who experienced treatment failure, treatment failure occurred within 2 years after treatment in 18 patients (81.8%) and within 5 years after treatment in 22 patients (100%). The 5-year OS, DFS, LC and DMFS rates for patients with MTV levels < 18.3 cm3 and ≥ 18.3 cm3 were 90.9% and 76.6% (p = 0.03), 83.5% and 67.7% (p = 0.051), 94.7% and 90.3% (p = 0.32), and 88.4% and 71.8% (p = 0.031), respectively.