First results of a phase III multi-centre randomized controlled trial of intensity modulated vs. conventional radiotherapy in head and neck cancer: PARSPORT (CRUK/03/005)
Reviewer: Christine E. Hill-Kayser, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 3, 2009
Presenter: Christopher Nutting Presenter's Affiliation: On behalf of the PARSPORT Trial Management Group, Cancer Research UK Type of Session: Scientific
Randomized trials have resulted in relatively good cure rates with combination therapy for cancers of the head and neck. Radiotherapy, either in the definitive or post-operative setting, is a mainstay of treatment for many head and neck cancers.
Unfortunately, radiation to the head and neck may be associated with significant late toxicity, including xerostomia, which may lead to poor dentition, speech/swallow dysfunction, and/or osteonecrosis.
Conventional radiotherapy, involving use of parallel opposed beams to large volumes, places the bilateral salivary glands at risk for damage that may result in moderate-severe xerostomia.
Intensity-modulated radiotherapy (IMRT) may allow sparing of one or both parotid glands, allowing maintenance of saliva production. Although phase II data has demonstrated benefits in salivary production with use of IMRT, outcomes have never been tested in a randomized setting.
The trial described here, PARSPORT, was designed to test the benefits of IMRT compared to conventional radiotherapy in treatment of squamous cell carcinomas of the head and neck.
Materials and Methods
Patients with cancers of the oropharynx/hypopharynx requiring radiotherapy were randomized to receive either conventional radiotherapy (CRT) or parotid-sparing IMRT.
Patients receiving post-operative radiotherapy without evidence of residual disease were planned to receive a total of 60 Gy in 30 fractions over 6 weeks.
Patients receiving definitive radiotherapy or post-operative radiotherapy with known residual disease were planned to receive 65 Gy in 30 fractions over 6 weeks.
All radiotherapy plans were reviewed by a centralized quality assurance committee, with each center submitting plans for approval before initiating patient treatment.
Dose-volume histogram data for tumor volumes and normal tissues were collected and maintained centrally.
Randomization was stratified by tumor site and treating institution; 94 patients from 6 centers were randomized to allow 84 assessable patients one year after treatment, with detection of a 30% difference in xerostomia rates with 90% power.
Patients were excluded from this study based on the following criteria:
The primary endpoint assessed was the incidence of grade 2-4 xerostomia (based on the LENTSOM scale) at one year after the end of therapy. Xerostomia grading was subjective, and defined as “partial but persistent or complete dryness.”
Secondary endpoints included acute and late radiation toxicity, overall survival, local control, quantitative saliva flow measurements, and quality of life.
Analysis was performed on an intent-to-treat basis, with analyses of toxicity endpoints performed on all patients for whom data were available.
94 total patients were randomized between the two treatment arms, 47 to receive CRT, and 47 to receive IMRT.
Of these, 72% were male.
Mean age was 58.4 years (range 37.5 – 82.8 years)
Median follow-up was 31.9 months
Primary disease sites were oropharynx in 85% of cases, and hypopharynx in 15%
AJCC disease stage was I/II in 23% of cases, and III/IV in 77%
Patient and tumor characteristics did not differ significantly between the two groups.
Radiotherapy was given as definitive treatment to 62% of patients receiving CRT, and 81% of those receiving IMRT.
Mean PTV radiotherapy dose was 65 Gy using both techniques.
Ipsilateral parotid mean dose was 59 Gy for patients receiving CRT, and 45 Gy for those receiving IMRT.
Contralateral parotid mean dose was 60 Gy for patients receiving CRT, and 26 Gy for those receiving IMRT.
Radiotherapy was given as post-operative treatment to 32% of patients receiving CRT, and 17% of those receiving IMRT.
Mean PTV radiotherapy dose was 64 Gy for patients receiving CRT, and 61 Gy for those receiving IMRT.
Ipsilateral parotid mean dose was 61 Gy for patients receiving CRT, and 50 Gy for those receiving IMRT.
Contralateral parotid mean dose was 57 Gy for patients receiving CRT, and 27 Gy for those receiving IMRT.
Neoadjuvant chemotherapy was delivered to 40% of patients receiving CRT, and 43% of those receiving IMRT.
Some acute toxicities, including dysphagia and dry mouth were encountered more by patients receiving CRT than IMRT (98% versus 87%, p = 0.05 and 91% versus 71%, p = 0.02, respectively). Fatigue was encountered more by patients receiving IMRT (76% versus 41% of patients receiving CRT, p < 0.01). Other acute toxicities, including clinical mucositis, pain, weight loss, and hair loss were similar between the two arms (acute toxicities were defines as those encountered during radiotherapy and up to 8 weeks after completion of radiotherapy).
At 12 months post-treatment, the incidence of LENT SOM grade 2 or greater xerostomia was significantly higher in patients who were treated with CRT versus IMRT (74% versus 39%, p = 0.004).
Differences were observed at three months post treatment (83% versus 62%, p = 0.04), and persisted at 6 and 18 months post treatment [86% versus 60% (p = 0.01) and 71% versus 29% (p = 0.003), respectively]
The incidence of RTOG grade 2 or greater xerostomia (defined as moderate or complete dryness of the mouth) was also consistently higher in patients treated with CRT versus IMRT. This was demonstrated at 3, 6, 12, and 18 months, and was significant at every measured time point.
Incidence of other RTOG toxicities of grade 2 or greater, including mucous membrane/subcutaneous tissue toxicity, joint toxicity, esophageal toxicity, bone, larynx, and spinal cord toxicity did not differ significantly between the two groups.
Overall survival was similar between the two groups [1 year overall survival 90.8% for patients treated with CRT versus 93.6% for those treated with IMRT, HR 1.05 (95% CI 0.38 – 2.90)].
Local-regional progression free survival (LRPFS) rates were also similar between the two groups [1 year LRPFS 88% for patients treated with CRT versus 87.3% for those treated with IMRT, HR 1.59 (95% CI 0.67 – 3.80).
The authors conclude that IMRT significantly reduces the risk of subjective xerostomia by about 50% for patients with pharyngeal cancers.
They note that acute radiation fatigue was more prevalent with IMRT than CRT, and comment that this may have been due to increased volume of brain irradiated with IMRT.
They note that studies with longer follow-up are warranted, but that the data presented here support adoption of IMRT as the standard of care for head and neck cancer patients.
The authors have performed a well-designed, well-executed study which is a valuable contribution to the literature.
Prior studies, both dosimetric and clinical (non-randomized) have demonstrated theoretical benefit to IMRT over CRT with regard to sparing of normal tissues during radiotherapy for cancers of the head and neck. Even so, the clinical impact of reduced parotid dose has remained somewhat undefined.
Given the profound impact of severe xerostomia on overall health and quality of life, prevention of long-term xerostomia is certainly important.
The authors demonstrate, in a randomized fashion, a benefit to parotid sparing IMRT in prevention of xerostomia during radiation for head and neck cancers. Based on their data, their conclusion that IMRT should be adopted as a standard of care is warranted.
Having said this, three caveats may be worth noting:
Firstly, all treatment plans executed as part of this study were subjected to quality assurance review. Certainly, IMRT is only as good as the radiation team designing IMRT plans, and IMRT technique will remain important in providing benefit to patients receiving it.
Secondly, some authors have voiced concerns regarding risk of “missed” treatment regions with use of IMRT for treatment of head and neck cancers. Certainly, use of IMRT does require improved patient immobilization and assessment of at-risk regions when compared to CRT. The follow-up in this study (one year) was not long enough to provide full assessment of local recurrence risk, nor overall survival between the two techniques. Longer follow-up with provide further data with regard to this, and will be of great interest.
Centers with significant experience in IMRT typically limit the dose to the parotid to lower levels then delivered in this study, which may result in even lower rates of xerostomia with IMRT.
Despite these limitations, the authors have performed an important study demonstrating benefit of IMRT in reducing xerostomia for head and neck cancer patients. At this point, this study supports adoption of IMRT as a standard of care, as the authors suggest, in centers with experience delivering IMRT for head and neck cancers. Further data on local recurrence and overall survival will be important, and should be publicized as it becomes available.
Jan 14, 2011 - Parotid-sparing intensity-modulated radiotherapy is less likely than conventional radiotherapy to result in dry mouth in patients treated for head and neck cancer, according to research published online Jan. 13 in The Lancet Oncology.