Moreover, this choice is in accordance with our belief that rectal bleeding is most strongly influenced by high dose levels (low n value) [20]. The 95% CI of the estimated TD50 and α/β parameters were established by the profile likelihood method as described by other authors [21]. All the calculations were performed by using the Matlab code (Release

6.5, The Mathworks Inc., Natick, Massachusetts). Results DVH analysis Differential and cumulative this website dose-volume histograms of each patient were collected. For both arms dose-volume constraints were well satisfied: for arm A, V50 and V70 resulted 38.3 ± 7.5% and 23.4 ± 5.5%, respectively; for arm B, V38 and V54 resulted 40.9 ± 6.8%. and 24.5 ± 4.4%, respectively (Fig. 1). From the small standard deviation of V50/V70 and V38/V54, it can be inferred that all patients were almost equally treated among each arm with respect to the dose distribution of the rectal wall. Figure 1 (a) The average with its standard deviation of the distribution of the cumulative rectal wall DVHs for the conventional arm. (b) The average with its standard deviation of the distribution of the cumulative rectal wall DVHs for the hypofractionated arm. To compare the two different treatment schemes, DVHs for the two arms have been both

normalized, converting the physical learn more dose in each volume fraction to the NTD2 (A.5) supposing an α/β ratio of 3 Gy. The plot in Fig. 2 shows together the TCL corrected DVHs for the two arms: the two curves are very close to each other, suggesting the equivalence of the conventional and the hypofractionated schemes in terms of the expected ≥ G2 late rectal toxicity. Figure 2 The averages of the distributions of the normalized cumulative rectal wall dose-volume-histograms

for arm A (dashed line) and for arm B (solid line). NTD2 on the X-axis indicates the biologically equivalent total dose normalized to the standard fraction of 2 Gy, supposing an α/β ratio of 3 Gy. Incidence of late toxicity The crude incidence ≥ G2 late rectal toxicity was 14.0% (8 patients) and 12.3% (7 patients) for the conventional and the hypo-fractionated arm respectively, after a median follow up of 30 months for both arms (range: 6-61 months for arm A, 6-63 months for arm B). In arm A, three patients experienced G3 toxicity and no patient developed G4; while in arm B no patients had late toxicity higher than G2. The actuarial ≥ G2 late toxicity at 30 months were 13.0% and 13.5% for arm A and B, respectively, as illustrated by the Kaplan-Meier curves in Fig. 3. No significant difference exists between the curves (p-value = 0.688 by the log rank test). Figure 3 Actuarial incidence of ≥ Grade 2 late rectal toxicity versus months after radiotherapy (mo.), for arm A and B.