Can combined volume and perfusion analysis improve the prediction of tumor control in cervical cancer?

Purpose: Quantitative analysis of either tumor volume or perfusion pattern using magnetic resonance imaging (MR) has been reported to be useful in the prediction of response to radiation therapy (RT) in cancer of the cervix. Since both analyses can be obtained in one single MR examination, the purpose of this study was to evaluate whether the combination of both analyses can further improve the accuracy in predicting tumor control after RT. Methods and Materials: Volume and perfusion analysis was performed prospectively in 20 patients with bulky carcinomas of the cervix: stages bulky 1B (2), IIB (6), IIIA (1), IIIB (9), IVA (1), and recurrent (1). Four studies were obtained in each patient: immediately before RT, after 20-22 Gy/∼2 weeks, after 40-45 Gy/∼4-5 weeks, and 4-8 weeks after therapy completion. The dynamic perfusion imaging was obtained at 3-second intervals in the sagittal plane over 120 seconds after rapid (9 ml/sec) intravenous injection of MR contrast agent (0.1 mmol/kg gadoteridol) using a power injector. Time/signal-intensity curves reflecting slope and relative signal intensity (rSI) of contrast enhancement in the tumor region were generated. Based on previous reports, low perfusion was defined as rSI<2.8, and high perfusion as rSI≥2.8. Tumor volumes were calculated by outlining the area of tumor in each slice on the T2-weighted images and multiplying by the slice profile. Median follow-up was 10 months (range: 3-24 months). Results: The combined analyses improved the prediction of local recurrence in high-risk groups to 75-80% from 60% by volume analysis alone, and 66% by perfusion analysis alon. This prediction can be made as early as 2 weeks (20-22 Gy) after start of RT. Particularly in medium-sized tumors, the prediction improved from 33% (i.e. similar to the overall incidence of 35%) to 80% (p=0.01; Table). For small tumors (<40 cm3), or high perfusion, the combined information did not improve outcome prediction. Conclusion: Based on these early data, the combination of tumor volume and perfusion analysis further improves the prediction of local failure compared to volume analysis or perfusion analysis alone, particularly in patients with medium-sized tumors