ICTR 2003: Translational research in clinics
Three-dimensional conformal vs. intensity-modulated radiotherapy in head-and-neck cancer patients: comparative analysis of dosimetric and technical parameters

Presented at ICTR 2003, Lugano, Switzerland, March 16–19, 2003.
https://doi.org/10.1016/j.ijrobp.2003.09.059Get rights and content

Abstract

Background and purpose

The use of intensity-modulated radiotherapy (IMRT) is now widely advocated for the treatment of head-and-neck cancers, to increase the therapeutic ratio of radiotherapy used as sole modality of treatment or in combination with chemotherapy. This report aims to summarize the technical and dosimetric factors to be taken into consideration to assess the respective advantages of the various high conformality treatments in radiotherapy, especially in the framework of quality assurance procedures.

Materials and methods

Twenty-six head-and-neck cancer patients were irradiated following a feasibility internal protocol with IMRT. Treatments were performed with either the static step-and-shoot (20) or the dynamic sliding window (6) techniques on a 6 MV Varian Clinac equipped with a multileaf collimator with 80 leaves. Dose plans were computed using commercial treatment planning systems: MDS-Nordion Helax-TMS for static cases and Varian Eclipse for dynamic cases. Dose plans were evaluated in terms of physical quantities based on dose–volume histograms and isodose distributions. Each IMRT plan was also compared to a reference 3D conformal therapy plan (3DCRT).

Results

Elective target volumes ranged from 530 to 1151 cm3 with a mean of 780 ± 141 cm3. Boost volumes ranged from 248 to 832 cm3 with a mean of 537 ± 165 cm3. Thirty-two dose plans were generated with static technique and 10 with dynamic. In the static mode, 6.8 ± 3.4 fields were applied on average with 12.5 ± 1.3 segments per field. In the static mode, 264 ± 56 MU per Gy were erogated, whereas in the dynamic mode, 387 ± 126 MU per Gy were erogated, to be compared to 147 ± 20 computed for reference 3DCRT plans. For all target volumes in general, conformity was improved compared to 3DCRT (e.g. V95 increased from 85% to 93% with p < 0.001, or equivalent uniform dose normalized to prescribed dose increased from 0.86 to 0.96 with p = 0.002). Irradiation of parotid glands or spinal cord improved, as well: For parotids, D2/3V reduced from 59 Gy to 41 Gy (p < 0.001). For spinal cord, Dmax reduced from about 40 Gy to about 30 Gy (p < 0.001).

Introduction

When, in May 2001, the Oncology Institute of Southern Switzerland (IOSI) launched its program of intensity-modulated radiotherapy (IMRT), an internal feasibility protocol was activated by the radiation physics unit of this institution, mainly for head-and-neck tumors. This program was designed to prove technical reliability and dosimetric quality of IMRT in the environment of a nonacademic, intermediate-sized center: The IOSI is indeed equipped with 2 linear accelerators (only 1 with IMRT capabilities at the time), 1 conventional simulator, and 3D planning systems. About 600 new patients are treated each year by a medical and radiation physics staff dimensioned for such a work load, but with minimal “buffer” for heavy escalation of routine activities.

The aim of IMRT programs is essentially to improve, within above-mentioned logistic constraints, the irradiation of organs at risk and healthy tissue without compromising target irradiation (1). Importantly enough, another potential role of IMRT for dose escalation in head and neck, ascertained at planning levels (2), can be to reduce overall treatment time and to increase local control without compromising acute or late side effects. The selection of anatomic sites to include in the IMRT program at IOSI was based on the following criteria: global improvement of treatment (head and neck), resolution of specific problems of conventional techniques (breast with internal mammary chain irradiation), resolution of particularly complex single cases (brain), and dose escalation (head and neck).

As a further development of a previous phase carried out, at the planning level, for various anatomic sites 1, 2, 3, 4, this report will focus on the main technical and dosimetric issues to take into consideration in an early phase of the IMRT clinical implementation, with particular attention to head-and-neck cancer treatment.

Section snippets

Patient selection, volume definition, dose prescriptions

Between May 2001 and the end of March 2003, 42 patients were treated with IMRT corresponding to 61 dose plans or target volumes. Twenty-six patients presented head-and-neck cancer: 18 received whole IMRT treatments, and 8 received mixed IMRT and three-dimensional conformal radiotherapy (3DCRT) treatments (5), for a total of 42 IMRT plans. Of those, 20 were treated with static step-and-shoot and 6 with dynamic sliding window delivery modes. Five patients were treated for breast cancer (4 of them

Technical summary of treated fields

In Table 2 a, summary of some technical aspects of the IMRT irradiations performed at IOSI for head-and-neck patients is reported. The numbers of beams per plan (i.e., number of different gantry angles) were 3 (2 plans), 5 (10 plans), 7 (30 plans), and 9 (2 plans); the average was 6.8 ± 3.4. In the SS mode, the range of the number of segments per beam was 9–15 and of the MUs per segment was 3–14.

Patients treated with IMRT were fit into conventional time slots of 15 min per session, and no

Dosimetric issues

Delivered IMRT dose plans showed a systematic and highly significant improvement in terms of target coverage compared to reference 3DCRT. In particular, there was a huge improvement in both the point and significant minimum doses and in V90 and V95. According to nEUD, the biologic impact is potentially large and quantified in about 10%. All these results are extremely encouraging, because they prove, on dosimetric grounds, the advantage for head-and-neck patients with IMRT, even compared with

Conclusion

Based on a 2-year clinical experience with photon IMRT, dosimetric data of a comparison at plan level between treated IMRT and reference 3DCRT dose plans show the superiority of IMRT in terms of both target coverage (minimum doses to target, V90, V95 or EUD estimates) and organs at risk (spinal cord and parotids) sparing; the data indicate that further improvement can be expected for salivary gland protection. Logistically and technically, the implementation of IMRT in clinical practice can be

Cited by (0)

The authors are grateful to Varian Medical Systems, International AG (Zug, Switzerland) for supporting a dedicated research program by providing the first clinical release of the Eclipse-HELIOS inverse treatment planning system. Part of the project was sponsored by the Swiss Federation against Cancer (OncoSuisse Grant No. 1151-09-2001).

View full text