Intensity-modulated radiation therapy A breakthrough treatment for nasopharynx cancer By Dr Lee Khai Mun, Ms Eileen Park, Dr Michael Back Nasopharynx cancer (NPC) is the sixth most common cancer affecting Singaporean males, and is a highly curable malignancy that is principally managed by definitive radiation therapy (RT). Although survival rates of greater than 90% are achievable for early stage disease, the RT doses required are high with surrounding normal tissues such as salivary glands, temporal lobe and optic nerves exposed to potential RT long-term side effects. Improvements in RT equipment over the past 10 years have resulted in major changes in RT delivery, which is now principally designed using CT scan imaging, and incorporates image-guided targeting of the tumours using three-dimensional conformal radiotherapy (3DCRT). Although it is superior in dose distribution over earlier two-dimensional (2D) traditional radiotherapy techniques, there is still significant irradiation of adjacent structures in the head and neck that impacts on quality of life in survivors.   RT-related side effects after NPC Therapy Whilst oncologists using 3DCRT specifically protect important dose-limiting structures such as the optic chiasm and spinal cord, there may be an unavoidable dose directed to tissues such as the inner ear, temporomandibular joints and temporal lobe. This may result in potential long-term damage of these structures giving rise to hearing loss, trismus and even temporal lobe necrosis.   However, the major predictable long-term effect of standard 3DCRT for NPC is reduced salivary flow or xerostomia resulting from unavoidable irradiation of the parotid salivary glands. The subsequent oral dryness impacts not only on speech and taste, but also compromises quality of life further by predisposing to oral fissures, ulcers, dental caries and even osteoradionecrosis of mandible. The degree of xerostomia is largely dependent upon the radiation dose and the volume of the salivary gland within the radiation field. Medications have only a minimal and short-term impact on this symptom and the most effective method of prevention is to reduce the radiation dose to the salivary glands.   Intensity-modulated radiation therapy Intensity-modulated radiotherapy (IMRT) represents a new and exciting paradigm in the field of radiation oncology. IMRT is superior to standard 3DCRT as it can deliver a more focused dose to the tumour and spare surrounding normal tissues. It achieves this more accurate radiation delivery by utilising improvements in modern radiation oncology machines, ie, linear accelerators and computer-driven calculation and delivery of radiation dose.   Multiple small radiation beams of varying shapes and sizes are used to match the contours of the tumour and spare the sensitive normal tissues. The optimum beam arrangements are designed by sophisticated inverse-planning software, which use the tumour and normal tissue information provided by the oncologist to calculate the best beam arrangement. Whilst standard 3DCRT may utilise three to five beam shapes to cover a tumour, an IMRT plan will often use more than 100 beamlets. With computer-driven delivery of the beams, the treatment time for the radiation patient is shortened, with the majority of treatments taking just 25 minutes each day.   Implementation of IMRT at The Cancer Institute (TCI) The potential therapeutic advantage of better dose conformation to tumour with improved sparing of normal tissues has been the main appeal for IMRT being implemented in radiation oncology units internationally. The use of inverse-planning technique with specific dose-constraints that the oncologist can direct to sensitive normal structures makes IMRT suitably exploitable for head and neck tumours, in particular NPC. Early international reports on the use of IMRT on NPC have demonstrated not only excellent tumour control rates, but also improvements in toxicity-profile. However, this treatment is highly sophisticated, resource-intensive and requires thorough quality assurance to achieve safe and accurate delivery.   Respecting this treatment complexity, TCI Radiation Oncology developed a 12-month phased programme from January 2004 for the development of IMRT at its principal unit at the National University Hospital (NUH). Not only was initial training undertaken in Europe and the US, but an overseas expert in the field from the University of California and San Francisco (UCSF) was invited with funding from Health Manpower Development Programme (HMDP) to train local staff in IMRT planning and dosimetry.   In addition, a transdisciplinary IMRT steering committee was established to develop the necessary treatment protocols and equipment acquisitions. Singapore’s Lee Foundation provided assistance in the purchase of necessary quality assurance hardware. A departmental survey was even conducted to assess and prepare staff’s readiness in the implementation of new technology.   Most importantly, from the outset, TCI decided that our early patient experiences commencing in February 2005 should be systematically recorded and analysed so as to be able to verify and evaluate the promised benefits of IMRT.     Comparison of IMRT and 3DCRT techniques A review of the radiotherapy plans and treatment-delivery techniques was performed for the first 20 patients treated radically with IMRT for NPC at The Cancer Institute. These IMRT plans and delivery techniques were compared with the radiotherapy plans for 10 other NPC patients treated with radical conventional 3DCRT just prior to the start of the IMRT programme.    The patients that underwent IMRT were fairly representative of NPC patients with thirteen of the IMRT and six of the 3DCRT having loco-regionally advanced disease and also treated with concurrent chemotherapy. Patients on IMRT were managed following the US Radiation Therapy Oncology Group protocol (RTOG-0225) with the prescribed dose of 70Gy in 33 fractions at 2.12Gy per fraction daily to the assigned planning target volume (PTV). Those receiving conventional 3D-CRT were treated with the prescribed total dose of 70.2 Gy in 39 fractions at 1.8 Gy per fraction daily to PTV (Figures 2a and 2b). TCI’s IMRT technique involved the use of seven to nine photon fields with 78-127 segments in a single phase, while the 3DCRT technique required 11 to 17 fields, including electron portals over two to three phases for each patient. As expected, the median treatment time per daily fraction was 26 minutes in contrast with 12 minutes for 3DCRT.   To determine the benefit of IMRT over 3DCRT techniques, the calculated radiation doses of the 30 patients’ tumours and specific normal tissues (temporal-lobes, temporomandibular joint and parotids) were assessed. The results outlining the median values for maximum RT dose (Gy) to these structures are shown in Figure 4 and demonstrate that with IMRT, the tumour can receive a higher radiation dose, whilst the normal tissues receive a significantly lower exposure. Specifically for the parotid glands, the maximum and mean doses were lower with IMRT than 3DCRT, and now below the levels that predict for the patient experiencing significant xerostomia.   Although long-term follow-up is needed for a full clinical evaluation of the impact of IMRT on the treatment of NPC with regards to tumour control and alleviation of side effects, like in earlier dosimetry studies our own early experience is very exciting. It has confirmed the superiority of IMRT over 3DCRT in better conformity of high-dose radiation to tumour and relative sparing of adjacent normal tissues, in particular the parotids, inner ears, temporomandibular joints, and temporal lobes. Logically, this should translate to better tumour control with reduction of radiation-related toxicities, thus enhancing the therapeutic index for the curative treatment of NPC.     Future directions with IMRT Improved conformity of the high dose-region to tumour also raises the potential for escalating the dose delivered to the tumour, with the hope of better tumour control especially for bulky or high-risk tumours. Indeed, apart from improved physical-dosimetry, IMRT also has the radio-biological advantage over conventional 3DCRT by allowing for “dose-painting” or preferential radiation dose-escalation in certain areas of the tumour volume that may have higher risk for relapse.   The relative sparing of side effects would be even more pertinent given the increased use of combined-modality therapy with radiation, such as concurrent chemotherapy and targeted therapies, which may have radio-sensitisation effects on normal tissues. This may allow therapies directed at potential metastatic sites to be added to treatment regimens more safely. Overall, IMRT techniques for NPC will be further targeting better cancer control whilst maintaining patients’ quality of life.