Dosimetric comparison and QA of 3D conformal and intensity modulated radiotherapy (IMRT) for para nasal sinus carcinoma

Full text: Treatment planning for para nasal sinus carcinoma (PNS) is quite complex for its close proximity to organs at risk (OARs) and the presence of air filled cavities near the tumour. 3D CRT and IMRT have reduced the doses to OARs significantly (1-3) compared to conventional techniques. In this study, an attempt has been made to obtain a class solution of 3D CRT plan that is comparable to an IMRT plan with respect to the tolerance doses to the OARs and PTV coverage. Plan parameters of 3D CRT were compared with IMRT. Dose delivery of 3D CRT was also verified using ion chambers and TLDs. Five patients with PNS who were treated with 3D CRT were studied. 3D CRT and IMRT plans were created for each patient. Planning CT of 5 mm slice thickness was obtained with the immobilization system in treatment position. Target volumes and OARs were drawn in accordance with ICRU 50. For 3D CRT plan, four 6 MV photon beams one anterior/vertex, two laterals, and a small anterior boost field with optimal wedge angles and weights were used. Bolus was used for tumours, involving skin. To reduce the dose to the eyes and the lens, eyes were shielded from the lateral beams. Hence there was a significant dose deficit near the region of maxillary sinus and posterior ethmoid. Since the final outcome for PNS tumours is a function of the control at the primary site, it is essential that the PTV be adequately covered by the prescription isodose. Hence to boost the dose to this region a small optimally weighted anterior field is added. Multi leaf collimators (MLCs) are shaped in such a way that in none of the beams, eyes and lens will be in the direct beam. Also small segments of MLC leaves are pulled in the beam to reduce the hotspots. The small anterior boost field where isocentre is shielded will have some uncertainty for its small size and more weight. Hence quality assurance was carried out, MLC shapes from the patient plan has been transferred to the head and neck phantom. Dose at the isocentre and at off axis, near the boost field were measured using ion chamber, and TLDs. IMRT plan is produced using seven coplanar beams placed 50 degree apart. Dose volume constraints were set, as follows: 95% of PTV should receive 95% of the dose; no volume of PTV should receive more than 107% of dose. The maximal tolerance dose to OARs are as follows: Spinal cord 45Gy, brain stem 54Gy, optic nerve 54 Gy, lens 6 Gy, and mean parotid dose 24Gy. Helios Optimization algorithm from Varian medical system is used for inverse planning. Dose prescription to PTV was 60 Gy in 30 fractions. For each patient, 3D CRT was compared with IMRT using isodose curves, dose volume histograms, and dose volume statistics. The minimal dose, maximal dose, percentage of PTV volume receiving 95% and 107%, V95% and V107%, dose conformity, homogeneity, maximum tolerance dose to OARs, dose to normal tissue were compared. There was no significant difference in the maximal dose (112-119%) and minimal dose (84- 92%) for 3D CRT and IMRT. For two patients in 3D CRT the minimal dose was as low as 75%, and it was found that the tumour was extending up to skin wherein in the build up region dose calculation was not very accurate. Mean V95% of PTV is 93% and 95%, mean V107% was 5% and 4% in 3D CRT and IMRT respectively. Both 3D CRT and IMRT were homogenous in PTV coverage, but conformity was superior in IMRT than in 3D CRT. IMRT doesn't offer any significant advantage of doses to critical structures. The normal tissue receives more dose in IMRT than in 3D CRT. V10%, percentage volume of normal tissue receiving 10% of the dose is 40% and 55%, V50% is 18% and 15% for 3D CRT and IMRT respectively. The dose measured for 3D CRT at isocentre and at off axis using both ion chamber, TLD is found to be with in ±4%. From the above results we conclude that with a few number of optimally placed beams with MLCs shielding the critical structures, 3D CRT can be an effective alternative to IMRT for PNS cancer to deliver 60Gy/30 fractions. IMRT plan uses large number of beams which m akes the treatment time longer. 3D CRT plan presented here consists of just four beams was faster to implement on linac, where time is a constraint. It also is less prone to error, and very much suitable to our country where most of the centers still do not have the infrastructure to treat patients with IMRT. Further the 3D CRT plan reduces the integral dose received by the normal tissue. Quality assurance results also confirm that the dose delivery is within ±4%. This 3D CRT plan has been accepted as an effective treatment modality for PNS tumours in our hospital. (author)