Metastatic site-adjustment improves predictive performance of metabolic tumor volume and total lesion glycolysis on FDG PET in patients with differentiated thyroid carcinoma.

Objectives For patients with differentiated thyroid carcinoma, metabolic tumor volume (MTV) and total lesion glycolysis (TLG) measured on FDG PET are useful parameters for predicting prognosis. Most studies investigating MTV and TLG considered all the metastatic lesions equally; i.e., whole-body MTV is calculated as the sum of the individual lesion’s MTV. In clinical practice, however, metastatic site is very important factor to affect patient prognosis. For instance, lung metastasis has a greater impact than neck node metastasis. In this context, the aim of this study was to measure MTV and TLG separately by metastatic site and to evaluate their performances. Furthermore, we propose new parameters of metastatic site-adjusted whole-body MTV and TLG.

Methods In this retrospective study, we investigated 299 patients with differentiated thyroid carcinoma who underwent FDG PET or PET/CT before receiving I-131 radioactive iodine therapy and had one or more FDG-avid metastatic lesion(s). After >6 hours fasting, the patients were injected with FDG (4.5MBq/kg), followed by whole-body scanning 1 hour after the injection. Images were analyzed as follows. All the uptake masses with SUV蠅3.0 were extracted automatically. Then, an experienced nuclear medicine categorized each tumor uptake as either neck node (n), distant node (d), lung (l), bone (b), or other organ/tissue (o) metastatic lesion. MTV was the tumor volume showing SUV蠅3.0, and TLG was tumor volume × SUVmean. Both MTV and TLG were recorded for each metastatic site (nMTV, dMTV, lMTV, bMTV, and oMTV; TLG likewise). To determine optimum cut-offs, any considerable cut-off values were exhaustively tested using Kaplan-Meier method with log-rank test, and the best cut-off producing the maximum chi-square value was adopted. The predictive performance of a parameter was evaluated using the maximized chi-square value.

Results All the 10 parameters (5 sites × MTV and TLG) reached statistical significance (P<0.0001) to distinguish patients between with good vs. poor prognosis when the best cut-off was used (Table 1), suggesting metastasis to any sites can affect patient prognosis. Here, using the best cut-off of each site’s MTV as a normalization factor, we introduced a new index of metastatic site-adjusted MTV as follows; newMTV = nMTV/11.74 + mMTV/0.77 + lMTV/1.58 + bMTV/15.97 + oMTV/2.89. The newMTV achieved better predictive performance than the ordinary whole-body MTV (i.e., the simple sum of each site’s MTV) (chi-square, 111.7 vs. 104.5). Similarly, the newTLG slightly improved the predictive performance of the ordinary whole-body TLG (chi-square, 107.2 vs. 106.8).

Conclusions MTV and TLG were measured separately by the metastatic site. Both MTV and TLG of any sites were significant predictor of patient prognosis. Metastatic site-adjusted MTV and TLG may have potential to improve the performance of simple whole-body MTV and TLG, although the normalization factors have room to optimize.

• Download figure
• Open in new tab
• Download powerpoint