Fax: (011) 31 20 6091278
Radiotherapy alone, versus radiotherapy with amifostine 3 times weekly, versus radiotherapy with amifostine 5 times weekly
A prospective randomized study in squamous cell head and neck cancer
Article first published online: 27 JUN 2006
Copyright © 2006 American Cancer Society
Volume 107, Issue 3, pages 544–553, 1 August 2006
How to Cite
Jellema, A. P., Slotman, B. J., Muller, M. J., Leemans, C. R., Smeele, L. E., Hoekman, K., Aaronson, N. K. and Langendijk, J. A. (2006), Radiotherapy alone, versus radiotherapy with amifostine 3 times weekly, versus radiotherapy with amifostine 5 times weekly. Cancer, 107: 544–553. doi: 10.1002/cncr.22020
- Issue published online: 18 JUL 2006
- Article first published online: 27 JUN 2006
- Manuscript Accepted: 24 MAR 2006
- Manuscript Revised: 10 MAR 2006
- Manuscript Received: 31 OCT 2005
- quality of life;
- sticky saliva
The main objective of this study was to investigate whether nondaily intravenous administration of amifostine was as effective as daily intravenous administration with regard to the reduction of the incidence of Grade 2 or greater xerostomia in patients with head and neck cancer.
Ninety-one patients who received bilateral irradiation for head and neck cancer were included. Thirty patients received no amifostine (AMI-0), 31 patients received amifostine at a dose of 200 mg/m2 3 times weekly (AMI-3), and 30 patients received amifostine at a dose of 200 mg/m2 daily (5 times weekly) (AMI-5). Acute and late xerostomia and quality of life (QOL) were assessed at baseline, 6 weeks later, and at 6-month intervals from 6 months to 24 months postradiotherapy.
Grade 2 or greater late xerostomia differed significantly at 6 months (AMI-0 74% vs. AMI-3 67% vs. AMI-5 52%; P = .03), but not thereafter. During follow-up, patient-rated xerostomia deteriorated more in AMI-0 patients (mean difference score:, 52 for AMI-0 compared with 25 for AMI-3, and 29 for AMI-5; P = .01). Nausea and emesis were reported most frequently as side effect, but Grade 2 or greater toxicity was observed in only 4 patients. However, 28% of patients discontinued amifostine before the end of radiotherapy.
Long-term, patient-rated xerostomia was less for the AMI-3 and AMI-5 groups through 2-year follow-up, but no difference was noted between the AMI-3 and AMI-5 groups. For late xerostomia according to the Radiation Therapy Oncology Group criteria, the same effect was observed at 6 months, but not thereafter. Cancer 2006. © 2006 American Cancer Society.
Radiotherapy is an important primary and adjuvant treatment modality for patients with head and neck cancer. Despite the beneficial effects with regard to locoregional tumor control, the radiation dose that can be administered is limited. Salivary dysfunction is an important side effect of radiotherapy and leads to xerostomia after irradiation in the head and neck area. The tolerance dose for damage to the parotid glands is approximately 26 to 40 grays (Gy).1–3 Below this dose, salivary dysfunction may be reversible. Higher doses may lead to irreversible damage of the salivary glands with permanent dryness, particularly in patients who receive doses that exceed 60 Gy.4–6
To increase the therapeutic ratio, it may be worthwhile to enhance the radioresistance of normal tissues without inducing protection of the tumor to radiation. Preclinical data supported the possible use of amifostine as radioprotector, because its radioprotective properties turned out to be greater in normal tissues compared with that in tumor cells.7
Amifostine is a thioorganic compound that originally was developed as a protector against radiation-induced toxicity in the event of nuclear war. Amifostine and its active metabolite, WR-1065, accumulate in many epithelial tissues, and the highest concentrations are found in the salivary glands, spleen, and kidneys.8 It has been reported that several normal tissues are protected, particularly the salivary glands. In tumor-bearing animals, there was selective cytoprotection of normal tissues from irradiation without protection of the tumor.9–11 Small clinical trials have suggested that amifostine protects against radiation-induced xerostomia and mucositis.12–14 Brizel et al.15 conducted a large, prospective, randomized multicenter trial for patients who received radiotherapy for head and neck cancer. Patients were randomized to receive radiotherapy alone versus amifostine (200 mg/m2 intravenously over 3 minutes) prior to radiotherapy. In their study, the incidence of Grade 2 or greater acute xerostomia decreased from 78% without amifostine to 51% with amifostine (P < .001), whereas the median dose at onset of xerostomia was 42 Gy without amifostine and 60 Gy with amifostine (P = .0001). After 2 years, no significant difference in survival or tumor control was observed.
In most studies concerning radioprotection, amifostine was used daily. The question arises whether nondaily administration of amifostine may be as effective as daily administration. A less frequent schedule may be important for the use of amifostine in clinical practice, because the administration of amifostine is both labor-intensive and costly. Several studies have supported the application of a nondaily scheme. First, Utley et al. demonstrated in mice that the half-life of WR-1065 in salivary glands after intravenous injection was significantly longer compared with that in other tissues.16 Second, in a randomized Phase II trial, 39 patients with squamous cell carcinoma of the head and neck received adjuvant or primary irradiation (60 Gy, 2 Gy per fraction, and 5 times weekly) with carboplatin on Days 1 through 5 and on Days 21 through 25.17 Twenty-five of those patients received 500-mg amifostine infusions. Although amifostine was administered only before carboplatin on Days 1 through 5 and Days 21 through 25, patients who received amifostine experienced a significant reduction in radiation-induced toxicity, such as xerostomia, mucositis, dysphagia, loss of taste, and dermatitis. Thus, although amifostine was given only during part of the treatment, a reduction in radiation-induced toxicity was observed.
Therefore, the primary objective of the current prospective, randomized Phase II study was to test the hypothesis that nondaily (3 times weekly) intravenous administration of amifostine is as effective as daily (5 times weekly) intravenous administration of amifostine with regard to the reduction of the incidence of Grade 2 or greater xerostomia compared with radiation alone.
MATERIALS AND METHODS
Patients were eligible for this study if they had Stage III/IVB (International Union Against Cancer, 199718) squamous cell carcinoma of the oral cavity, oropharynx, hypopharynx, and/or larynx or if they had lymph node metastases in the head and neck area from an unknown primary treated with bilateral primary or postoperative radiotherapy with curative intent, a minimal life expectancy of 12 months, and a World Health Organization performance score from 0 to 2. Patients were required to have a good understanding of the Dutch language to be able to complete the quality-of-life (QoL) questionnaires. Furthermore, at the time of randomization, 75% of the parotid gland volume was expected to receive a radiation dose of at least 40 Gy. This was confirmed by using computed tomography-assisted treatment planning and dose-volume histograms. Excluded were patients with distant metastases (M1), previously irradiated patients, patients who were treated in combination with induction or concurrent chemotherapy, and patients with tumors that originated in the salivary glands. Pregnant patients, patients who were participating in another investigational trial, and patients in poor general or psychological conditions that did not permit them to complete the trial also were excluded. Because of the potential occurrence of hypotension after the administration of amifostine, patients who had severe cerebrovascular disease, poor renal function, or sustained hypotension not secondary to antihypertensive medication also were excluded. Written informed consent was obtained from all patients. The study was approved by the Medical Ethical Committee of the Vrije Universiteit University Medical Center and was in accordance with the Helsinki Declaration.
This was a prospective, randomized Phase II study comparing radiotherapy alone with radiotherapy in combination with 2 different administration schedules of amifostine. Patients were randomized to receive radiotherapy alone (AMI-0), radiotherapy in combination with amifostine at a dose of 200 mg/m2 5 times weekly (AMI-5) before irradiation, or radiotherapy in combination with amifostine at a dose of 200 mg/m2 3 times weekly (AMI-3) on Monday, Wednesday, and Friday before irradiation. In patients who received accelerated radiotherapy (6 times weekly with a second fraction on Fridays), amifostine was given only before the first fraction on that day. Random assignment of patients was performed at the Department of Radiation Oncology of the Vrije Universiteit University Medical Center by using a permuted block design. Patients were stratified by primary tumor site (oral cavity vs. oropharynx vs. others), radiotherapy (primary vs. postoperative), and submandibular glands removed (yes vs. no).
Administration of Amifostine
Amifostine was supplied in single, 10-mL vials that contained 500 mg per vial. The lyophilized dosage was stored in a refrigerator (from 2°C to 8°C). Sterile 0.9% sodium chloride solution was used for reconstitution. Half an hour before administration of amifostine, all patients were hydrated with 250 cc fluid orally. Amifostine was administered by slow intravenous injection over 3 to 5 minutes (200 mg/m2) 15 to 30 minutes before irradiation, because it had been reported that administration in several minutes produced less acute toxicity compared with injection over 15 minutes.19 During the administration of amifostine and for 15 minutes after administration, patients were kept in supine position. Blood pressure was checked immediately before, immediately after, and 10 minutes after the intravenous administration of amifostine. Prophylactic antiemetic premedication was prescribed. Notations of all adverse events were made whether or not they were related to treatment.
Details regarding radiation therapy have been reported previously.20 In brief, all patients were treated with megavolt equipment using isocenter techniques after 3-dimensional planning. Generally, the initial target volume was irradiated by using 2 opposing lateral fields with an anterior field to cover the lower jugular and supraclavicular lymph node areas. All patients received 46 Gy to electively treated areas; boost doses varied from 56 Gy (in patients who had negative surgical margins) to 63.5 Gy (in patients who had lymph node metastasis with extranodal spread or positive margins). Patients who were treated primarily with radiotherapy received 70 Gy to macroscopic tumor.
The primary endpoint of this study was Grade 2 or greater late radiation-induced xerostomia according to the Radiation Therapy Oncology Group/European Organization for Research and Treatment of Cancer (RTOG/EORTC) Late Radiation Morbidity Scoring Scheme.21 The secondary endpoints were Grade 2 or greater, acute, radiation-induced xerostomia; patient-rated xerostomia; and sticky saliva; QoL; locoregional tumor control and survival. Late radiation-induced xerostomia was assessed before radiotherapy and at 6 months, 12 months, 18 months, and 24 months after the completion of radiation. Acute radiation-induced xerostomia was scored according to the RTOG Acute Radiation Morbidity Scoring Criteria before radiotherapy, on a weekly basis during radiotherapy, and up to 6 weeks after radiotherapy.
Assessment of QoL and patient-rated xerostomia
For the evaluation of QoL, the EORTC Core Questionnaire (QLQ-C30, version 3.0) was used with the supplemental head and neck-specific module, the QLQ-H&N35.22, 23 To assess self-reported xerostomia, sticky saliva, and other head-and-neck symptoms, the EORTC QLQ-H&N35 was used.23 Patients rated their symptoms on a 1 to 4 scale (no symptoms, mild symptoms, moderate symptoms, and severe symptoms), which was converted linearly to a 0 to 100 metric. Higher scores represented a greater degree of symptoms.24
The questionnaire was administered before the start of treatment (baseline), 6 weeks later, and at 6-month intervals from 6 months to 24 months after the completion of radiation. Patients were asked to participate in the study during the first visit at the Radiotherapy Department. After informed consent was obtained, the questionnaires were distributed to the patients, who were asked to fill in the questionnaire at the same time. During regular follow-up visits, a research nurse asked patients to fill in the questionnaires before they visited the physician. When a patient did not comply with the planned visit, the questionnaire was sent by mail. When the questionnaire was not returned within 4 days, the patient was contacted by telephone and was asked to complete and return the questionnaire.
For the purpose of the current study, the effect of amifostine on the self-reported symptom scales for xerostomia and sticky saliva were considered most important. The other QoL dimensions were analyzed on an exploratory basis.
Locoregional tumor control and survival
Locoregional control (LRC) was defined as no tumor recurrence above the clavicles within the irradiated area. LRC and overall survival were calculated from the first day of radiotherapy.
Based on the results from the study by Brizel et al.,15 it was expected that the incidence of Grade 2 or greater xerostomia would be reduced from 75% to 50% when radiotherapy was combined with amifostine (AMI-5). A similar reduction rate was expected from the AMI-3 compared with AMI-0. With a 70% power to detect such a difference at the 5% significance level, 30 patients in each arm were required. The analysis was performed according to the intention-to-treat principle.
To test differences between treatment arms regarding the primary endpoint, i.e., the frequency of Grade 2 or greater late radiation-induced morbidity, the chi-square test was used. In the univariate analysis, LRC and overall survival were estimated with the Kaplan–Meier method. To test the statistical significance of differences between curves, the log-rank test was used.
Changes in symptoms and QoL items were evaluated with a repeated-measures analysis of variance by using a mixed-effect modeling procedure (SAS Proc. Mixed). In contrast to a “complete cases analysis,” the mixed-effects model retains in the analysis patients who drop out during follow-up. Trends over time for dropouts and complete cases are estimated under the assumption that patients within the same group have the same pattern over time. SAS Proc. Mixed uses the method of restricted maximum likelihood to estimate the parameters of the model. The F test was used to test for main effects of group and time and for the interaction effect of group × time.
The study was open for inclusion from August 1999 to August 2003. Ninety-one patients were included, with 31 patients each in the AMI-0 and AMI-5 arms and 30 patients in the AMI-3 arm. The demographic and tumor characteristics of the population are listed in Table 1. No significant differences were noted between the 3 treatment arms. Sixty-seven percent of the patients were male. The mean age at the start of radiotherapy was 55 years (range, 24–73 years). Nearly 50% of the tumors originated in the oropharynx (46%). The majority of patients received postoperative irradiation (78%).
|Variable||AMI-0 (n = 31)||AMI-3 (n = 30)||AMI-5 (n = 30)||Total (n = 91)||P|
|Type of radiation||NS|
|Lymph node classification||NS|
The mean dose to the parotid glands was 43.5 Gy (95% confidence interval [95% CI], 37.3–49.7) in the AMI-0 arm, 49.2 Gy (95% CI, 45.9–52.5) in the AMI-3 arm, and 47.9 Gy (95% CI, 44.8–52.0) in the AMI-5 arm and did not differ significantly (P = .173). The mean dose to the submandibular glands was 57.3 Gy (95% CI, 54.4–60.2) in the AMI-0 arm, 59.9 Gy (95% CI, 56.6–63.3), in the AMI-3 arm, and 55.9 Gy (95% CI, 52.3–59.5) in the AMI-5 arm and also did not differ significantly (P = .162). No differences were observed regarding the mean dose-volume histograms in the parotid and submandibular glands between the 3 treatment arms (Fig. 1). Two patients were treated off protocol. One patient developed lymphangitis carcinomatosa on the second day of treatment, and the other patient developed pulmonary metastases during radiotherapy. All other patients completed radiotherapy as planned.
Side Effects and Compliance of Amifostine
Among the patients who received amifostine, minor side effects were observed (Table 2). Nausea and emesis were reported most frequently during radiotherapy and occurred more frequently among patients who received amifostine (P = .004). Although there was a trend toward more nausea and emesis in the AMI-5 arm compared with the AMI-3 arm, the difference was not significant (P = .183). Other, less frequently reported side effects were hypotension and allergic skin reactions. Hypotension often was mild and of short duration. In total, 24 patients (28%) discontinued amifostine before the completion of radiotherapy (Table 2), and there was no difference observed between the AMI-3 arm and the AMI-5 arm. Nineteen patients (21%) discontinued amifostine before they received 40 Gy of radiotherapy. Of the 24 patients who discontinued amifostine, the most frequently reported reason was nausea and emesis (15 patients). Other reasons for discontinuing amifostine were allergic skin reactions in 4 patients, hypotension (not amifostine related) in 2 patients, tumor progression in 1 patient, difficulties in accessing the vein in 1 patient and refusal in 1 patient.
|Side Effect||AMI-0(n = 31)||AMI-3(n = 30)||AMI-5(n = 30)||Chi-Square P|
Late radiation-induced xerostomia at 6 months, 12 months, and 24 months
At 6 months, the scores for late xerostomia according to the RTOG criteria differed significantly for the 3 study arms (P = .03). In the AMI-0 arm, 74% of patients reported Grade 2 or greater late xerostomia compared with 67% of patients in the AMI-3 arm and 52% of patients in the AMI-5 arm (Table 3). In the AMI-5 arm, 15% of patients had no xerostomia, which was not the case in the other 2 treatment arms. After 12 months, no significant differences were noted between the 3 treatment arms. Furthermore, no significant differences were noted when patients in the AMI-0 arm were compared with all patients who received amifostine (the AMI-3 and AMI-5 arms).
|Treatment arm||No. of patients (%)|
|6 months||12 months||18 months||24 months|
|AMI-0||20/27 (74)||15/22 (68)||13/20 (65)||11/19 (58)|
|AMI-3||18/27 (67)||13/22 (59)||8/16 (50)||8/14 (57)|
|AMI-5||15/29 (52)||15/27 (56)||12/21 (57)||11/15 (73)|
|P (df = 2)||.03||.24||.59||.81|
Acute radiation-induced reactions
At baseline, no significant differences were observed between the 3 arms with regard to scores for xerostomia according to RTOG criteria. Xerostomia was present in 3% of patients in the AMI-0 and AMI-3 arms and in 13% of patients in the AMI-5 arm (nonsignificant). In all 3 treatment arms, xerostomia increased to Grade 2 at the end of treatment (AMI-0, 80% of patients; AMI-3, 77% of patients; and AMI-5, 73% of patients). The time of onset of acute xerostomia did not differ significantly between arms (Fig. 2A). No significant differences were observed between the 3 treatment arms with regard to acute toxicity of the mucous membranes and pharynx (Fig. 2B and 2C).
Changes in patient-rated xerostomia and sticky saliva
At baseline, no significant differences were noted with regard to patient-rated xerostomia or patient-rated sticky saliva. During follow-up, the deterioration of patient-rated xerostomia among patients in the AMI-0 arm was significantly worse compared with that among patients in the AMI-3 and AMI-5 arms (Fig. 3A) (P = .016).
The addition of amifostine did not have any effect on patient-rated sticky saliva (Fig. 3B). No significant differences between the 3 treatment arms were noted with regard to the other QoL dimensions (data not presented).
Survival and LRC
The 2-year LRC rate was 79% for patients in the AMI-0 arm, 67% for patients in the AMI-3 arm, and 83% for patients in the AMI-5 arm (P = .31) (Fig. 4A). The overall survival rate at 2 years was 70% for patients in the AMI-0 arm, 84% for patients in the AMI-3 arm, and 58% for patients in the AMI-5 arm (P = .26) (Fig. 4B).
To our knowledge, this is the first study to report on the effects of 2 different regimens of amifostine in addition to radiotherapy. The results indicate that the addition of amifostine 5 times weekly to radiotherapy temporarily reduced the incidence of late radiation-induced xerostomia according to the RTOG/EORTC criteria at 6 months, but not thereafter. The schedule of amifostine 3 times weekly appeared to be less effective with regard to the incidence of late radiation-induced xerostomia. However, with regard to patient-rated xerostomia, significantly less deterioration was observed when amifostine was added to radiotherapy at all time points tested, and no difference was noted between the AMI-3 and AMI-5 treatment arms.
A number of authors have reported on the application of amifostine in patients with head and neck cancer who are treated with radiotherapy.13, 14, 17 To our knowledge the most important study on this topic published to date was reported by Brizel et al.15 In that study, patients who were treated for head and neck cancer were assigned randomly to receive (postoperative) radiotherapy alone versus (postoperative) radiotherapy with amifostine 5 times weekly. A significant reduction of acute and late Grade 2 or greater xerostomia was observed as well as a significantly greater unstimulated salivary gland output when amifostine was added to radiation. Moreover, patients who received amifostine experienced a better clinical benefit, which was measured by using a patient-benefit questionnaire.25
In the current study, we observed no difference in acute xerostomia between the study arms. This may be explained by the finding that virtually all patients experienced serious sticky saliva and loss of taste, 2 factors that influence the rating of acute xerostomia according to the RTOG criteria. It remains unclear whether loss of taste and sticky saliva were taken into account in the assessment of acute xerostomia in the study by Brizel et al.15
A number of authors found that postradiotherapy salivary gland function depended significantly on the dose distribution in the parotid glands. One of the main shortcomings of other studies that reported on the radioprotective properties of amifostine was that the amount of salivary tissue included in the radiation portals was estimated based on the simulation films and not on 3-dimensional information derived by planning computed tomography scans. In the current study, the dose distribution in the salivary glands was evaluated by using dose-volume histograms, and no differences were observed regarding the mean dose and mean dose-volume histograms in the parotid and submandibular glands between the 3 treatment arms.
Most grading systems of xerostomia and endpoints in studies that assess dry mouth are based on functional disabilities.21, 26, 27 However, in clinical practice, xerostomia is primarily a QoL issue. A lack of concordance between objective endpoints of xerostomia, such as salivary flow and xerostomia-related questionnaires assessing patient-rated xerostomia, has been reported.5, 28–30 In addition, the physician-rated RTOG toxicity score system, although it is used widely, has a potential lack of sensitivity, because it has not been validated prospectively. Therefore, patient-rated xerostomia was considered the most important endpoint.
In the current study, the EORTC QLQ-H&N35 was used to assess patient-rated xerostomia and sticky saliva. It is important to note that this instrument contains only 1 question regarding xerostomia and 1 question regarding sticky saliva using 4-point Likert scales. Subjective xerostomia is a complex symptom, which may vary during the day and may be more pronounced in a given situation, such as speaking, eating, and physical activity. Therefore, QoL scales with few specific questions may not be sensitive enough. Recently, Eisbruch et al.30 described a validated xerostomia questionnaire that contains 8 questions on xerostomia rated on an 11-point Likert scale. Such validated questionnaires that address different aspects of xerostomia and xerostomia-related symptoms may enable the detection of more subtle changes compared with the methods that were used in the current studies.
Although patients who received amifostine in the present study fared significantly better with regard to patient-rated xerostomia than patients who received radiotherapy alone, this difference did not translate into better scores on more general QoL dimensions or on head and neck symptoms (data not shown). These results indicate that amifostine mainly has a radioprotective effect on salivary gland tissue and not on other normal tissues, such as mucosa. This finding is supported by the lack of significant group differences for acute mucositis and acute toxicity of the pharynx. Furthermore, the results indicate that prevention of xerostomia does not necessarily translate into improvement in the general QoL for patients, something that often has been suggested but never has been confirmed empirically.
One of the problems identified in the current study was the reduced compliance because of the toxicity of amifostine, in particular, nausea and emesis. Although toxicity scores were in the lower range and were considered mild, for many patients, nausea and emesis complaints were a serious burden, even after adjustment of the antiemetic regimen. Oral correction of hydration status did not reduce nausea sufficiently in all patients. The toxicity from amifostine in the current study was comparable to that reported by Brizel et al., who observed that 47% of amifostine-treated patients had any grade of nausea, 37% had any grade of emesis, 15% had hypotension, and 5% had allergic reactions. Those results were reflected in a noncompletion rate of 22%.15 Others also have reported reduced compliance because of the toxicity of amifostine.13–15, 31 The level of treatment discontinuation varied from 11% to 41%, depending on the dose of amifostine administered and the radiation scheme applied. One study was ended preliminary because of toxicity.31 Therefore, even though there have been reports of preventing xerostomia with the intravenous administration of amifostine preirradiation, toxicity is an important obstacle in clinical practice. A decrease in the administration time of amifostine by rapid intravenous push may reduce toxicity further.32 In addition, delivering amifostine subcutaneously may be promising; however, even using that route, a number of patients (11%) reported to suffer from substantial toxicity, and compliance was reduced by 10% to 35%.33, 34 The current trial was started as a Phase II study with the intention to proceed to a Phase III study. However, because of smaller than expected differences between the treatment arms, the possibility of tumor protection,35 and the difficulties with compliance, eventually, we decided not to proceed with a Phase III study. Furthermore, the question arises whether there is a place for radioprotective agents like amifostine in the era of new technologic developments in radiation oncology, such as intensity-modulated radiation. A number of studies have indicated that the probability of radiation-induced xerostomia depends primarily on the mean parotiddose1, 2 and/or the mean submandibular dose20 and that radiation-induced xerostomia can be prevented by the use of these newer techniques. Determining whether amifostine may be of use in a specific subgroup of patients, depending on the mean dose to the salivary glands, will require further investigation.
In conclusion, despite only moderate levels of compliance, amifostine offered temporary protection from late radiation-induced, observer-rated xerostomia at 6 months and had a longer lasting effect on patient-rated xerostomia, as measured during 2 years of follow-up. The late radioprotective properties of amifostine on a schedule of 3 times weekly appear to be similar to those of amifostine on a schedule of 5 times weekly. However, toxicity may hamper implementation in clinical practice. Investigation of alternative administration techniques, adequate hydration, and more antiemetic regimens may lead to reduced acute toxicity and may increase compliance.
Amifostine kindly was provided by Schering Plough.
- 8Clinical experiences with a chemical radioprotector in tumour radiotherapy: WR-2721. In: SugharaT, editor. Modification of Radiosensitivity in Cancer Treatment. Tokyo: Academic Press; 1984: 64–81..
- 18Head and neck tumors. In: SobinLH, WittekindC, editors. TNM Classification of Malignant Tumors. 6th ed. New York: John Wiley & Sons; 2002: 19–47., .
- 24EORTC Quality of Life Study Group. EORTC QLQ-C30 Scoring Manual. Brussels: EORTC; 1995., , , , on behalf of the
- 27National Institutes of Health. Common Toxicity Criteria for Adverse Events, version 3.0. Available at: http://ctep.info. nih.gov/CTC3/ctc.htm. Accessed April 1, 2003.
- 32Amifostine (Ethyol) given by rapid IV push is safe and tolerable. Proc Am Soc Clin Oncol. 2001; 20: 300b., , .