Intensity-modulated radiotherapy significantly reduces the risk of subjective xerostomia by about 50% in patients with pharyngeal tumors, according to the first results of the multicenter, phase III PARSPORT trial.
Cancer Research UK’s PARSPORT (Parotid-Sparing Intensity-Modulated Radiation Therapy Compared With Conventional Radiation Therapy in Treating Patients With Oropharyngeal or Hypopharyngeal Cancer Who Are at High Risk of Radiation-Induced Xerostomia) trial evenly randomized 94 patients with pharyngeal tumors to conventional radiotherapy (conventional radiotherapy ) or intensity-modulated radiotherapy (IMRT). A three-dimensional technique, IMRT produces highly conformal dose distributions that can reduce the radiation dose to the salivary glands and normal tissue.
At 12 months, the incidence of grade 2 or higher xerostomia was 74% in CRT patients vs. 39% in IMRT patients, based on the subjective portion of the LENT/SOM (Late Effects on Normal Tissue-Subjective/Objective Management) questionnaire (P = .004).
The benefit of IMRT appeared to continue over time, with an 18-month xerostomia incidence of 71% with conventional therapy vs. 29% with IMRT (P = .003), principal investigator Dr. Christopher Nutting reported in a late-breaking abstract presentation at the annual meeting of the American Society of Clinical Oncology.
A similar pattern was observed using the RTOG (Radiotherapy Oncology Group) scale. The incidence of at least grade 2 xerostomia was 64% with CRT vs. 41% with IMRT at 12 months (P = .05), and 81% vs. 20% at 18 months (P less than .001).
This is the first randomized IMRT trial in head and neck squamous cell carcinoma, although phase II data suggest that parotid gland-sparing IMRT maintains saliva production, said Dr. Nutting, head of the head and neck unit at the Royal Marsden Hospital in London.
Grade 2 or higher radiation fatigue during and up to 8 weeks post treatment was significantly higher in the IMRT arm than in the CRT arm (76% vs. 41%; P less than .01), possibly because of more irradiation of brain tissue, he said. No other significant differences in acute or late toxicities were seen between the two arms.
“These data support the adoption of IMRT as the standard of care for head and neck cancer patients,” Dr. Nutting said.
The importance of the PARSPORT trial is that it confirms findings from two prior randomized trials in nasopharyngeal carcinoma patients (who also have a high risk of xerostomia), and it “shows that an improvement in radiation technology can translate into a decrease in toxicity,” Dr. Anthony Cmelak, medical director of the Vanderbilt-Ingram Cancer Center at Franklin (Tenn.), said during a discussion of the study.
He applauded the study’s use of a centralized quality assurance program because, he said, there are no set standards of delivery or quality assurance for IMRT in the community. Radiotherapy was delivered according to protocol in 43 CRT patients and 46 IMRT patients. (The remaining five patients either refused or were ineligible for treatment, or deviated from the protocol.).
When given as primary treatment, the mean dosage in both arms was 65 Gy in 30 fractions over 6 weeks. The mean contralateral parotid dose was 60 Gy in the CRT arm, which is capable of producing long-term damage, compared with 26 Gy in the IMRT arm, Dr. Nutting said. Ipsilateral parotid mean doses were similar, at 59 Gy vs. 45 Gy.
When given postoperatively, the mean radiotherapy dose was 64 Gy in the CRT arm vs. 61 Gy in the IMRT arm, the mean contralateral parotid dose was 57 Gy vs. 27 Gy, and the mean ipsilateral parotid dose was 61 Gy vs. 50 Gy.
In all, 85% of patients had tumors arising in the oropharynx and 15% in the hypopharynx; 77% had AJCC (American Joint Committee on Cancer) stage III/IV disease, and about 40% received neoadjuvant chemotherapy. Their mean age was 58 years.
Dr. Nutting said that one of the concerns about not treating part of the anatomy of the head and neck region is that locoregional progression-free survival may be poorer with IMRT. No significant difference was observed in this outcome between the IMRT (87.3%) and CRT (88%) arms (hazard ratio, 1.59).
With a median follow-up of 32 months, 91% of CRT and 93.6% of IMRT patients were alive at 1 year, although the confidence interval of 0.38-2.90 (HR, 1.05) precludes any significant conclusions, he said.
Dr. Cmelak said outstanding questions are whether the use of concurrent chemotherapy that sensitizes tissue would influence xerostomia outcomes or the ability of IMRT to spare parotid parenchyma, and whether reduced xerostomia justifies the increased integral dose. It has been asserted that the large number of beams and monitor units used in IMRT leads to an increase in integral dose (that is, the total amount of energy absorbed by a patient during radiation).
He said that the future of radiation delivery lies in arc-modulated radiation technology, which is available on roughly 70% of machines being produced today and which can reduce the integral dose by about one-half and a 15-minute IMRT session to about 5 minutes.
Finally, Dr. Cmelak said that additional research should include a cost-benefit comparison, citing a substantial difference in Medicare global costs of $23,715 for IMRT vs. $12,850 for 3-D radiation. Quality-of-life data from PARSPORT will be forthcoming.
The investigators disclosed receiving honoraria from Elekta AB and Varian Medical Systems Inc.