Search

Aims To determine the influence of dose and fractionation on tumour characteristics, toxicity, disease control and survival outcomes in T1 glottic carcinoma. Materials and methods Between 1975 and 2000, treatment charts of 652 patients with T1 glottic carcinoma who received curative radiation with four hypofractionated schedules (50 Gy/15 fractions [3.3 Gy/fraction] or 55 Gy/16 fractions [3.43 Gy/fraction] or 60 Gy/24 fractions or 62.5 Gy/25 fractions [2.5 Gy/fraction]) were analysed. The patients were divided into two groups based on fraction size 3 Gy. Local control and overall survival were calculated. Patient- and tumour-related factors affecting local control were analysed using univariate and multivariate analysis. Factors affecting late toxicity were also analysed. Results The local control and overall survival at 10 years were 84 and 86.1%, respectively, for T1 glottic carcinoma. The response to radiation had a significant effect on local control with univariate analysis ( P = 0.001). Other factors, such as beam energy, anterior commissure involvement and fractionation, did not affect local control. Persistent radiation oedema was seen in 123 patients (23.4%) and was significantly worse in patients who received radiation with a larger field size (>36 cm 2 ) on a telecobalt machine ( P Conclusions Radical radiotherapy schedules incorporating a higher dose per fraction yield acceptable local control rates and late toxicity. Telecobalt therapy for early glottic cancer is a safe alternative to treatment with 6 MV photons on a linear accelerator in terms of local control and late toxicity as long as field sizes smaller than 36 cm 2 are used.

The dosimetric outcomes of radiograph- and computed tomography-based plans generated with various optimisation strategies were compared using dose volume indices for partial breast intraoperative implants.Eighteen patients with early stage breast cancer underwent conventional orthogonal radiograph and computed tomography to generate dosimetric data. Catheter reconstruction and delineation of the lumpectomy cavity, the planning target volume (PTV) and the ipsilateral breast were carried out on computed tomography images. For each patient, geometrically optimised plans (P(xray) and P(CT)) were generated using the active loading length based on the PTV defined from radiographs (PTV(xray)) and computed tomography (PTV(CT)). The plan P(CT) was further optimised graphically and yielded P(graphical). Plans were compared using the coverage index (CI), the external volume index (EI), the dose homogeneity index (DHI), the overdose volume index (OI) and the conformal index (COIN).The mean CI of the lumpectomy cavity estimated from P(xray), P(CT) and P(graphical) was 0.80, 0.82 and 0.92, respectively. The corresponding values for PTV(CT) were 0.69, 0.71 and 0.85. P(graphical) showed an increase in CI by 23 and 19% with respect to P(xray) and P(CT) (P0.001 in all) with a decrease in DHI from 0.81 to 0.71 (P0.001) and increase in OI from 0.041 to 0.087 (P0.001). The EI was highest in P(xray) (mean 44 cm(3)) as compared with 25 cm(3) in P(CT) and 30 cm(3) in P(graphical). A significant improvement in COIN was observed in P(graphical) (mean 0.68) compared with P(xray) (0.48) and P(CT) (0.58) (P0.001).Objective dosimetric evaluation on three-dimensional computed tomography confirms its superiority over conventional two-dimensional radiograph-based planning in terms of a reduction in normal breast irradiated with the prescription dose and improvement in conformity. Interactive graphical optimisation based on the target volume in computed tomography further improves conformity with a reduction in dose homogeneity. The use of dose volume indices allows the comparison of different plans and can be used as a tool to correlate dosimetric with clinical outcome.