Radiotherapy Elsevier RADION
and Oncology,
18 (1990) 107-l 16
107
00689
Accelerated fractionation D. S. Lamb,
radiotherapy neck cancer
N. A. Spry, A. J. Gray,
A. D. Johnson,
for advanced head and S. R. Alexander
and M. J. Dally
Wellington Hospital, Wellington, New Zealand
(Received 6 April 1989, revision received 16 November
Key words: Radiation
therapy; Accelerated
1989, accepted
fractionation;
8 December
1989)
Head and neck cancer
Summary Between 1981 and 1986, 89 patients with advanced head and neck squamous cancer were treated with a continuous accelerated fractionation radiotherapy (AFRT) regimen. Three fractions of 1.80 Gy, 4 h apart, were given on three treatment days per week (Monday, Wednesday, Friday), and the tumour dose was taken to 59.40 Gy in 33 fractions in 24-25 days. Acute mucosal reactions were generally quite severe, but a split was avoided by providing the patient with intensive support, often as an in-patient, until the reactions settled. Late radiation effects have been comparable to those obtained with conventional fractionation. The probability of local-regional control was 47% at 3 years for 69 previously untreated patients, whereas it was only 12% at one year for 20 patients treated for recurrence after radical surgery. Fifty-eight previously untreated patients with tumours arising in the upper aero-digestive tract were analysed in greater detail. The probability of local-regional control at 3 years was 78% for 17 Stage III patients and 15 % for 3 1 Stage IV patients. This schedule of continuous AFRT is feasible and merits further investigation.
The optimal treatment for advanced (Stage III and IV) squamous carcinoma arising in the head and neck region has yet to be determined. Many of these patients are inoperable, or refuse radical surgery. Unfortunately, conventionally adminisAddress for correspondence:
Oncology, Wellington New Zealand. 0167-8140/90/$03.50
Dr. D. S. Lamb, Department of Hospital, Private Bag, Wellington,
0 1990 Elsevier Science Publishers
tered radiotherapy has yielded poor survival rates in most series [4-6,191 and this has led to the investigation of a number of alternative approaches designed to increase the radiation effect. Hyperbaric oxygen [ 1,3,13,14] and hypoxic cell sensitisers [ 6,26,29] have yielded generally disappointing results, with the small improvement in some series outweighed by technical difficulties in administration and increased toxicity respectively. More recently, there has been interest in the use of synchronous chemotherapy and radiation
B.V. (Biomedical
Division)
108 [ 9,12,15,20]. An alternative approach has been the use of unconventional fractionation regimes. A radical course of radiotherapy is conventionally given over 6 to 7 weeks using a single fraction of 1.80-2.00 Gy per day, on 5 treatment days per week. This schedule was largely developed empirically, with the 24-h interval between fractions being selected principally for patient and staff convenience. There is little radiobiological evidence to suggest that such timing is optimal in terms of the therapeutic ratio. Proliferation of clonogenic tumour cells may be significant during a 6 to 7-week course of radiotherapy and could lead to ultimate failure in some cases. Animal studies have shown that the cell kinetics of a tumour alter after irradiation, with a decrease in the potential doubling time and a decrease in the cell loss factor [2,28]. Various workers have calculated the clonogenic cell number doubling time (TclO,) in human tumours during a course of radiotherapy by comparison of the isoeffective dose for tumour control with different overall treatment times [ 17,21,22]. They found that Tclon approximated much more closely to the potential doubling time before treatment, rather than the volume doubling time before treatment. T C,onfor tumours of the head and neck region lies between 4 and 7 days. Accelerated fractionation radiotherapy (AFRT), whereby the overall treatment time is reduced by giving 2 or 3 conventional sized fractions on each treatment day, should therefore decrease the number of clonogenic cells that must be killed by the radiation [ 8,311. Shortening of the overall treatment time can be expected to increase the acute radiation reaction, as the population of clonogenic normal epithelial cells will be more drastically reduced in the short term. In contrast, late radiation effects should not be significantly worse, as these are largely dependent on the fraction size and total dose, providing there is an adequate time interval between fractions for repair of sub-lethal damage in normal tissues [ 241. The patients in this study were treated with a schedule comprising 3 fractions of 1.80 Gy, 4 h apart, on three treatment days per week (Monday,
Wednesday, Friday), and the tumour dose was taken to 59.40 Gy in 24-25 days. This regimen is a modification of the one described by Gonzales et al. [lo], who gave 3 fractions of 1.80 Gy on 5 days per vweek, and a tumour dose of 48.60-54.00 Gy in 11-12 days. When this regimen was tried in Wellington the acute reactions were felt to be unacceptably severe. In addition, the tumour had little time to regress during treatment, which might not have allowed for reoxygenation. Our modification enabled both the overall time and the tumour dose to be increased.
Patients and methods Between 1981 and 1986, 89 patients with advanced head and neck squamous cell carcinoma localised above the clavicles were treated with AFRT at Wellington Hospital. All patients gave their informed consent according to the standards of the institution. The early patients in this series were part of a pilot study which has already been reported [ 111. Sixty-nine patients received radiotherapy as primary treatment for their malignancy. Most of these patients were considered unsuitable for surgery because of advanced tumour, other significant medical conditions or old age. Twenty patients received radiotherapy for recurrent tumour following initial treatment with radical surgery. There were almost three times as many male as female patients (66 : 23). The median age of all patients was 62 years with a range of 12 to 85 years. Twenty-eight of the 69 previously untreated patients were aged 70 years or more. Site of tumour is shown in Table I. The TNM classification (UICC) of the 69 previously untreated patients is shown in Table II. The Stage Grouping of previously untreated patients with tumours of the oral cavity, pharynx and larynx is shown in Table III. Prior to treatment, patients underwent examination under anaesthetic by a head and neck surgeon, with a radiation oncologist in attend-
109 TABLE
TABLE
I
Stage grouping of previously untreated
Site of tumour. Number recurrent
Number previously untreated
5
Oral cavity Oropharynx Hypopharynx Nasopharynx Larynx glottic supraglottic Nasal cavity Skin Unknown primary Lip Parotid Total
TABLE
III
31 9 6 4
1 2 2
I
2 1
11 3 12 3 2 1
20
89
I -
69
All patients
II
TNM classification
of 69 previously untreated
NO
N,
T, T, TX T‘l TX”
1 9 10 14 3
2 3 2 10
Total
31
21
N2
TO
All patients M,. a No T classification
patients.
N3
Total
1 1 2
3 1 2 2 3
3 4 15 15 29 3
4
11
69
for nasal cavity.
ante. The diagnosis of squamous cell or undifferentiated carcinoma was confirmed histologically in all cases. Haematological and biochemical screening and chest X-ray were mandatory, and the majority of patients also had a CT scan of the neck and involved region. Dental assessment was routinely performed. All patients were treated on a Varian 4 MeV linear accelerator. Using a schedule comprising 3 fractions of 1.80 Gy, 4 h apart, on three treatment days per week (Monday, Wednesday,
Site
patients. Total
Stage I
II
III
IV
Oral cavity Oropharynx Hypopharynx Nasopharynx Larynx glottic Larynx supraglottic
1 -
2 1 -
6 3 2
18 5 4 2
26 9 5 4
1
4
._
5
5
2
2
9
Total
1
9
17
31
58
Note: There is no stage grouping for skin, nasal cavity, or unknown primary.
Friday), the tumour dose was taken to 59.40 Gy in 33 fractions in 24-25 days. Patients were treated using a shrinking field technique. Radiotherapy was initially to a large volume which included the regional lymph nodes as well as the primary tumour with a margin. Parallel opposed lateral fields were used to treat the primary tumour and upper neck, and a direct anterior field (with mid-line shielding) the lower neck. The tumour dose was taken to 37.80 Gy in 21 fractions in 15 days, with the tumour dose defined as the mid-line dose on the central axis for the parallel opposed lateral fields and the dose at 20 mm depth in the supraclavicular fossae for the direct anterior field. The minimum tumour dose was not permitted to fall below 90% of the maximum dose for the parallel opposed lateral fields. Tissue compensators were not used. The dose for subclinical disease was kept to 37.80 Gy because of uncertainty about the long term tolerance of normal tissues, particularly the spinal cord, to accelerated fractionation. Frail elderly patients did not receive prophylactic irradiation of the whole neck, as the risk of recurrence from untreated microscopic metastatic disease was outweighed by the possibility of increased morbidity from a larger treatment volume.
110
After 37.80 Gy the treatment volume was reduced to include only detectable tumour with a margin, and the spinal cord was excluded. A further 21.60 Gy in 12 fractions in 8 days was then delivered to the tumour. The fields used were either parallel opposed laterals or a technique using antero- and postero-laterals on the ipsilateral side and a direct lateral on the contralateral side. The tumour dose for the small volume was taken as the maximum dose, and the minimum tumour dose was not permitted to fall below 93% of the maximum dose. anaesthetic All patients used antiseptic mouthwash and candida prophylaxis on a regular basis. Patients who developed significant side effects were admitted to hospital to ensure that symptoms were controlled satisfactorily and that adequate nutrition was maintained. Severe mucositis was not considered a reason for suspending radiotherapy. Early in the study, 12 patients were treated with chemotherapy prior to radiotherapy. All received two or three courses of &-platinum, vinblastine and bleomycin, except for one patient with very bulky neck nodes from an unknown primary who was given high dose melphalan followed by autologous marrow rescue. Nine patients had a partial response, but there were no complete responders. Of the nine previously untreated patients who were given chemotherapy, there were six partial responders. Sixteen of the 69 previously untreated patients underwent surgery after radiotherapy for persistent tumour, relapse or radiation complications. Follow-up was calculated from the start of radiotherapy in all cases, and life-table analysis of data was performed according to the method described by Peto [25]. Complete response (CR) was defined as the absence of any clinical evidence of cancer 3 months after start of radiotherapy. Multivariate analysis of data was performed using the step-wise proportional-hazards model.
Results The median follow up of the 89 patients was 44 months, with a range of 14 to 83 months, Local-regional
control
This is defined as the control of cancer at the primary site and within the cervical lymph nodes. Figure 1 shows that for all patients, the probability of local-regional control was 39% at 3 years. However, local-regional control for the 69 previously untreated patients was 47% at 3 years, whereas for the 20 patients treated for recurrence after radical surgery local-regional control was only 12% at one year (p < 0.001). The CR rate in the 69 previously untreated patients was 68%. In this group there were 33 local-regional failures - one case of nodal relapse outside the radiation field, two cases of nodal relapse within the prophylactic field and 30 cases of persistent or recurrent tumour at sites of presenting disease. Figure 2 shows the probability of local-regional control according to stage for 58 previously untreated patients with tumours of the oral cavity, pharynx or larynx. The whole group had a CR rate of 7 1 y0 and a 3 year local regional control rate of
Fig. 1. Probability of local-regional control previous treatment status.
according
to
111 Distant metastases Of the 69 previously untreated patients, nine developed distant metastases, but only two of these did not have associated local-regional failure. Survival
Fig. 2. Probability of local-regional control according to stage for patients with previously untreated tumours of oral cavity, pharynx or larynx.
49%. There was a 100% CR rate for the 27 patients with Stage I, II or III tumours, whereas the CR rate for the 31 patients with Stage IV tumours was only 45 % . The probability of local-regional control was strongly dependent upon stage (p < O.OOOl), and at 3 years was loo%, 78% and 15% for Stage I + II, Stage III and Stage IV tumours, respectively. Table IV shows the results of multivariate analysis of risk factors for local-regional failure in the 58 previously untreated p.atients with tumours of the oral cavity, pharynx or larynx. Only the T and N staging reached statistical significance, whereas age, sex, site and grade of tumour did not.
TABLE
There were 56 deaths in the 89 patients treated, of which 50 were due to cancer. The remaining six patients died of intercurrent disease (three patients of myocardial infarction, one of a cerebra-vascular accident, one of a gastrointestinal haemorrhage and one of a motor vehicle accident) and were free of clinical evidence of cancer at the time of death. Figure 3 shows that the probability of survival from cancer for the 69 previously untreated patients was 44% at 3 years, whereas the probability of survival for the 20 patients treated for recurrence after surgery was only 20% at 2 years (p < 0.01). Figure 4 shows the probability of survival from cancer according to stage for 58 previously untreated patients with tumours arising in the oral cavity, pharynx or larynx. The probability of survival was strongly dependent on stage
PRO$ABLlTY
IV
Multivariate
analysis of risk factors for local-regional failure.
Factor
P
T,T,T3T,
< 0.000 1
&N&N, Site: Oral cavity/pharynx/larynx Grade: l/2/3,4 Age (yrs): < 65/> 65 Sex: M/F
0.03 0.73 0.68 0.25 0.12
Fig. 3. Probability
of survival according ment status.
to previous treat-
dysphagia due to the mucositis was strongly dependent on the site of the primary tumour. Patients with tumours of the oral cavity or pharynx often required admission during the third week of treatment to ensure that adequate nutrition was maintained. In contrast, patients with tumours of the larynx and other primary sites were usually able to complete treatment as outpatients. In all, 46 of the 89 patients (52%) required admission, and the mean duration of admission was 18 days. 28 patients (31%) required treatment with total parenteral nutrition (TPN) because oral intake was inadequate and their weight continued to fall despite opiates and a variety of topical treatments. The mean duration of TPN when used was 12 days. Skin reactions were much less severe than the mucosal reactions, probably due to the skin sparing effect of the megavoltage radiation. When the skin sparing effect was lost due to the surface contour, localised moist desquamation usually developed by the end of treatment to the high dose volume (week 4). Otherwise, moderate to marked erythema with some dry desquamation was the standard maximal reaction. A degree of xerostomia occurred in all patients, and developed during the second week of treatment. However, it only became severe if both parotid salivary glands were included in the high dose volume, and then exacerbated the dysphagia due to mucositis. Acute radiation reactions were slightly exaggerated in 11 of the 12 patients who received prior
Fig. 4. Probability of survival according to stage for patients with previously untreated tumours of oral cavity, pharynx or larynx.
(p < 0.0001) and at 3 years was 90x, 75% and 15% for Stage I and II, Stage III and Stage IV tumours, respectively. The 12 patients treated with initial chemotherapy had a similar survival to patients treated with radiotherapy alone. Complications Acute The acute mucosal reactions were generally severe. Table V shows the mean mucosal reaction according to time and accumulated dose for 21 recently treated patients who had mucosa included in the high dose volume. The severity of
TABLE
V
Mean mucosal reaction according to accumulated
dose and time.
Week Accumulated dose (Gy) Mucosal reaction (mean score)a
0
1
2
3
4
0
16.2
32.4
48.6
59.4
0
1.1
2.8
4.1
4.1
a Modified from Wang [30]. 0, Normal appearance; 1, mild erythema; mucositis.
2, severe erythema;
3, studded
mucositis;
5
6
8
10
3.4
2.6
1.4
0.6
4, confluent
mucositis;
5, haemorrhagic
113 chemotherapy. However, in one patient who was initially treated with high dose melphalan and autologous marrow rescue, the reaction was markedly exaggerated. This patient with neck nodes from an unknown primary developed prolonged grade 5 mucositis and extensive moist desquamation of the skin. He subsequently developed severe late complications. There were three deaths attributed to acute toxicity in the 89 patients, with two cases of bronchopneumonia and one of septicaemia. For statistical purposes, these were considered as cancer deaths. All toxic deaths occurred in patients with recurrent or Stage IV tumours. One patient had received prior treatment with chemotherapy. Late In this report, late complications are those persisting or developing 3 months or more after completion of radiotherapy.
Other late complications. One patient with a T,, N, carcinoma of the supraglottic larynx required total laryngectomy for radionecrosis 7 months after completion of radiotherapy. Interestingly, at the time of surgery he was found to be hypothyroid. Another patient with a T,, N, carcinoma of the oral cavity has moderately severe trismus 38 months after treatment. Salvage surgery Sixteen of the 69 previously untreated patients underwent surgery following radiotherapy. Two patients required surgery for radionecrosis, and three had excision of suspicious neck nodes which histologically showed no cancer. The other 11 patients had limited excision of persistent or recurrent cancer and two of these have since remained free of cancer.
Discussion Mucosa. Asymptomatic slight to moderate mucosal oedema persisted in 14 patients for up to 8 months after radiotherapy. It was more often a feature in advanced tumours of the larynx. Slight to moderate xerostomia has persisted in 18 patients with oropharyngeal or oral cavity tumours when one or both parotid salivary glands were included in the high dose volume. Skin and subcutaneous tissues. One patient developed localised skin necrosis 22 months after radiotherapy, which necessitated full thickness skin grafting of the affected area. This patient with neck nodes from an unknown primary was treated with high dose melphalan and autologous marrow rescue prior to radiotherapy, and also experienced exaggerated acute radiation effects. Two other patients developed asymptomatic subcutaneous fibrosis in the anterior neck. In both, a marked variation in surface contour resulted in dosimetry problems which might have been avoided by using tissue compensators. Satisfactory healing of the skin and other tissues occurred in all 16 patients who underwent surgery after radiotherapy.
As expected, the acute complications of our regimen of AFRT were considerably more severe than those of conventional fractionation, with mucositis a particular problem. However, we decided against modtication of our treatment regimen and opted instead for a policy of intensive support to tide the patient over the acute reaction. This necessitated the admission of 52% of patients, compared to our previous experience of 2 1% for conventionally fractionated radiotherapy (66.00 Gy in 33 fractions over 45 days). Patient management became easier as experience of our AFRT regimen increased and the timing of mucosal reactions could be accurately predicted. The three toxic deaths all occurred in the early part of the study, and although these patients all had very advanced tumour, it is possible their deaths might have been avoided if their problems had been anticipated. Later our policy was to readily admit patients with extensive tumours of the oral cavity and pharynx when the mucosal reaction became severe, and to initiate parenteral feeding as soon as oral intake became inadequate.
114 This approach permitted treatment of even quite elderly patients, and 28 of the previously untreated patients were aged 70 years or more. The late complications of our AFRT regimen have been quite acceptable, and not obviously worse than our previous experience with conventional fractionation. Of the 69 previously untreated patients, only three developed moderate to severe late complications, with one case each of laryngeal necrosis, skin necrosis and trismus. Our incidence of laryngeal necrosis is 7 % (one of 14 patients), identical to the 5-12x incidence reported in the literature for conventional fractionation [ 71. Healing was not a problem in those patients who underwent surgery after AFRT. Our results support the widely held belief that any treatment used for recurrence is much less likely to be effective than the same treatment given ab initio. Also, the probability of local-regional control and survival are similar at any given time, confirming that survival is largely dependent on achieving local-regional control [ 181. Our 3 year local-regional control rate of 47%
TABLE
for all previously untreated patients is comparable to recently published results for patients treated with conventionally fractionated radiotherapy, where local-regional control rates of 26-&l% were achieved [ 4,6,29]. Our previously untreated patients with tumours of the upper aero-digestive tract were analysed in greater detail as they comprise a less heterogeneous population compared to the group as a whole. We used the Stage Grouping System in order to obtain sub-groups with sufhcient numbers for statistical comparison. The 3 year localregional control rate was 78% for 17 Stage III tumours and 15% for 31 Stage IV tumours. The figure for Stage III tumours is encouraging, but there are to few patients and follow-up is too short for any firm conclusions. Table VI summarises other reported studies of AFRT for head and neck cancer. It should be noted that many studies are not pure AFRT, as either a reduction has been made in the total dose, or there has been no appreciable reduction in the overall treatment time because a split has been
VI
AFRT for head and neck cancer. Author
No. of patients
Dose/F (Gy)
No. of fractions per day
Total dose (Gy)
Time (wks)
Split (wks)
Results
CR at 8 months 68% CR at 3 months 86% 3-Year survival 44% 2-Year actuarial local control rate 65% Highly significant increase in 3-year actuarial local control rate (68% vs.
Perracchia and Salti [23] Svoboda [27]
22 59
2 1.75-2.3
3 3
48-54 50-55
-1.5 -2
-
Knee et al. [ 161
53
1.2-2.0
2
- 70-74
6
-
321
1.6
2
64-67.2
6
2
303 a 308
1.8 1.6
1 3
-65 67.2-72
7.2 6-7
3-4
159b
2
1
7
-
Wang [30]
46%) Van den Bogaert and van der Schueren [29]
a Historical controls. ’ Randomised controls.
70
No difference in 3-year actuarial local control rate (35% for all patients)
115 introduced [ 241. These compromises have been made-in an attempt to reduce the acute normal tissue reactions. The only randomised study comparing AFRT to conventional fractionation has been performed by the EORTC [29], and this showed no significant difference in the 3 year local control rate, which was 35 % for all patients. However, the overall treatment time for AFRT with a split and conventional fractionation was almost identical. Wang [ 301 also employed a split in his schedule, but still reduced the overall treatment time by over a week. His 3 year localregional control rate for 321 patients was 61 y0 compared to 46% for historical controls. Knee and colleagues [ 161 at the M.D. Anderson Hospital used the concomitant boost technique, a variation of AFRT whereby the boost is delivered as a second daily fraction during the basic treatment course. Their regimen reduced the total duration of treatment by about 2 weeks. They treated a heterogeneous group of 53 patients with a tumour dose of 70-74 Gy. The local-regional control rate at 2 years was 65 %, but there was a relatively high incidence of moderate to severe late complications. Our continuous AFRT schedule is feasible and merits further investigation. There is now enough accumulated experience in AFRT for advanced head and neck cancer to justify a multi-centre randomised study comparing a continuous schedule of AFRT with conventional fractionation.
Acknowledgements The authors would like to thank Dr. P. J. Dady, Mr. J. Hornibrook, Mr. P. Blake, Mr. N. Thomson, Mr. D. Palmer and Mr. J. Scott, for their help in the management of these patients. They would also like to thank Miss S. Parker for typing the manuscript and Dr. P. B. Roberts for his advice.
References 1 Berry, G. H., Dixon, B. and Ward, A. J. The Leeds results for radiotherapy in H.B.O. for carcinoma of the head and neck. Clin. Radiol. 30: 591-592, 1979. 2 Denekamp, J. Cell Kinetics and Cancer Therapy. C. C. Thomas, Springfield, IL, 1982. 3 Denham, J. W., Yeoh, E. K., Wittwer, G., Ward, G. G., Ahmad, A. S. and Harvey, N. D. M. Radiation therapy in hyperbaric oxygen for head and neck cancer at Royal Adelaide Hospital - 1964 to 1980. Int. J. Radiat. Oncol. Biol. Phys. 13: 201-208, 1987. 4 Duncan, W., Orr, J. A., Arnott, S. J., Jack, W. J. L., Kerr, G. R. and Williams, J. R. Fast neutron therapy for squamous cell carcinoma in the head and neck region: results of an randomised trial. Int. J. Radiat. Oncol. Biol. Phys. 13: 171-178, 1987. 5 Fazekas, J. T., Sommer, C. and Kramer, S. Adjuvant intravenous methotrexate or definitive radiotherapy alone for advanced squamous cancers of the oral cavity, oropharynx, supraglottic larynx or hypopharynx: concluding report of an RTOG randomised trial on 638 patients. Int. J. Radiat. Oncol. Biol. Phys. 6: 533-541, 1980. 6 Fazekas, J., Pajak, T. F., Wasserman, T., Marcia& V., Davis, L., Kramer, S., Rotman, M. and Stetz, J. Failure of misonidazole sensitised radiotherapy to impact upon outcome among stage III-IV squamous cancers of the head and neck. Int. J. Radiat. Oncol. Biol. Phys. 13: 1155-l 160, 1987. I Ferguson, B. J., Hudson, W. R. and Farmer, J. C. Hyperbaric oxygen therapy for laryngeal radionecrosis. Ann. Otol. Rhinol. Laryngol. 96: l-6, 1987. 8 Fowler, J. F. Potential for increasing the differential response between tumors and normal tissues: can proliferation rate be used? Int. J. Radiat. Oncol. Biol. Phys. 12: 641-645, 1986. 9 Fu, K. K., Phillips, T. L., Silverberg, I. J., Jacobs, C., Goffinet, D. R., Chun, C., Friedman, M. A., Kohler, M., McWhirter, K. and Carter, S. K. Combined radiotherapy and chemotherapy with bleomycin and methotrexate for advanced inoperable head and neck cancer: update of a Northern California Oncology Group randomised trial. J. Clin. Oncol. 5: 1410-1418, 1987. 10 Gonzales, D., Breur, K. and van der Schueren, E. Preliminary results in advanced head and neck cancer with radiotherapy by multiple fractions a day. Clin. Radio]. 3 1: 417-421, 1980. 11 Gray, A. J. Treatment of advanced head and neck cancer with accelerated fractionation. Int. J. Radiat. Oncol. Biol. Phys. 12: 9-12, 1986. 12 Haselow, R. E., Adams, G. S., Oken, M. M., Gondsmit, A., Lerner, H. J. and Marsh, J. C. Cis-platinum (DDP) with radiation therapy (RT) for locally advanced unresectable head and neck cancer. Proc. Am. Sot. Clin. Oncol. 2: 160, 1983.
116 13 Henk, J. M., Kunkler, P. B. and Smith, C. W. Radiotherapy and hyperbaric oxygen in head and neck cancer. Lancet II: 101-103, 1977. 14 Henk, J. M. and Smith, C. W. Radiotherapy and hyperbaric oxygen in head and neck cancer. Interim report of 2nd clinical trial. Lancet II: 104-105, 1977. 15 Kaplan, M. J., Hahn, S. S., Johns, M. E., Stewart, F. M., Constable, W. C. and Cantrell, R. Mitomycin and fluorouracil with concomitant radiotherapy in head and neck cancer. Arch. Otolaryngol. 111: 220-222, 1985. 16 Knee, R., Fields, R. S. and Peters, L. J. Concomitant boost radiotherapy for advanced squamous cell cancer of the head and neck. Radiother. Oncol. 4: l-7, 1985. 17 Maciejewski, B., Gunhild, P-B. and Trott, K-R. The influence of the number of fractions and of overall treatment time on local control and late complication rate in squamous cell carcinoma of larynx. Int. J. Radiat. Oncol. Biol. Phys. 9: 321-328, 1983. 18 Merino, 0. R., Lindberg, R. D. and Fletcher, G. H. An analysis of distant metastases from squamous cell carcinoma of the upper respiratory and digestive tracts. Cancer 40: 145-151, 1977. 19 Million, R. R. and Cassisi, N. J. Management of Head and Neck Cancer. J.B. Lippincott Company, Philadelphia, 1984. 20 Murthy, A. K., Taylor, S. G., Showel, J., Caldarelli, D. D., Hutchinson, J. C., Holinger, L. D., Witt, T. R. and Hoover, S. Treatment of advanced head and neck cancer with concomitant radiation and chemotherapy. Int. J. Radiat. Oncol. Biol. Phys. 13: 1807-1813, 1987. 21 Niederer, J., Hawkins, N. V., Rider, W. D. and Till, J. E. Failure analysis of radical radiation therapy of supraglottic laryngeal carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 2: 621-629, 1976. 22 Parsons, J., Bova, F. J. and Million, R. R. A re-evaluation of split-course technique for squamous cell carcinoma of the head and neck. Int. J. Radiat. Oncol. Biol. Phys. 6: 1645-1652, 1980.
23 Peracchia, G. and Salti, C. Radiotherapy with thrice-aday fractionation in a short overall time. Clinical experiences. Int. J. Radiat. Oncol. Biol. Phys. 7: 99-104, 1981. 24 Peters, L. J. and Ang, K. K. Unconventional fractionation schemes in radiotherapy. In: Important Advances in Oncology, pp. 269-286. Editors: V. J. De Vita, S. Hellman and S. A. Rosenberg. J.B. Lippincott Company, Philadelphia, 1986. 25 Peto, R., Pike, M. C., Armitage, P., Breslow, N. E., Cox, D. R., Howard, S. V., Mantel, N., McPherson, K., Peto, J. and Smith, P. G. Design and analysis of randomised clinical trials requiring prolonged observation of each patient. II. Analysis and examples. Br. J. Cancer 35: l-39, 1977. 26 Seely, R. and Cridland, S. The treatment of locally advanced head and neck cancer with misonidazole, hyperbaric oxygen and irradiation. Int. J. Radiat. Oncol. Biol. Phys. 10: 1721-1723, 1984. V. H. J. Accelerated fractionation: the 27 Svoboda, Portsmouth experience 1972-1984. In: Proceedings of Varian’s Fourth European Clinac Users Meeting, Malta, May 25-26, pp. 70-75. Varian, Zug, Switzerland, 1984. 28 Trott, K-R. and Kummermehr, J. What is known about tumour proliferation rates to chose between accelerated fractionation or hyperfractionation? Radiother. Oncol. 3: l-9, 1985. 29 Van den Bogaert, W. and van der Schueren, E. Early results of the EORTC randomised clinical trial on multiple fractions per day (MFD) and misonidazole in advanced head and neck cancer. Int. J. Radiat. Oncol. Biol. Phys. 12: 587-591, 1986. 30 Wang, C. C., Blitzer, P. H. and Suit, H. Twice a-day radiation therapy for cancer of the head and neck. Cancer 55: 2100-2104, 1985. 31 Withers, H. R. Biologic basis for altered fractionation schemes. Cancer 55: 2086-2095, 1985.
and Oncology,
18 (1990) 107-l 16
107
00689
Accelerated fractionation D. S. Lamb,
radiotherapy neck cancer
N. A. Spry, A. J. Gray,
A. D. Johnson,
for advanced head and S. R. Alexander
and M. J. Dally
Wellington Hospital, Wellington, New Zealand
(Received 6 April 1989, revision received 16 November
Key words: Radiation
therapy; Accelerated
1989, accepted
fractionation;
8 December
1989)
Head and neck cancer
Summary Between 1981 and 1986, 89 patients with advanced head and neck squamous cancer were treated with a continuous accelerated fractionation radiotherapy (AFRT) regimen. Three fractions of 1.80 Gy, 4 h apart, were given on three treatment days per week (Monday, Wednesday, Friday), and the tumour dose was taken to 59.40 Gy in 33 fractions in 24-25 days. Acute mucosal reactions were generally quite severe, but a split was avoided by providing the patient with intensive support, often as an in-patient, until the reactions settled. Late radiation effects have been comparable to those obtained with conventional fractionation. The probability of local-regional control was 47% at 3 years for 69 previously untreated patients, whereas it was only 12% at one year for 20 patients treated for recurrence after radical surgery. Fifty-eight previously untreated patients with tumours arising in the upper aero-digestive tract were analysed in greater detail. The probability of local-regional control at 3 years was 78% for 17 Stage III patients and 15 % for 3 1 Stage IV patients. This schedule of continuous AFRT is feasible and merits further investigation.
The optimal treatment for advanced (Stage III and IV) squamous carcinoma arising in the head and neck region has yet to be determined. Many of these patients are inoperable, or refuse radical surgery. Unfortunately, conventionally adminisAddress for correspondence:
Oncology, Wellington New Zealand. 0167-8140/90/$03.50
Dr. D. S. Lamb, Department of Hospital, Private Bag, Wellington,
0 1990 Elsevier Science Publishers
tered radiotherapy has yielded poor survival rates in most series [4-6,191 and this has led to the investigation of a number of alternative approaches designed to increase the radiation effect. Hyperbaric oxygen [ 1,3,13,14] and hypoxic cell sensitisers [ 6,26,29] have yielded generally disappointing results, with the small improvement in some series outweighed by technical difficulties in administration and increased toxicity respectively. More recently, there has been interest in the use of synchronous chemotherapy and radiation
B.V. (Biomedical
Division)
108 [ 9,12,15,20]. An alternative approach has been the use of unconventional fractionation regimes. A radical course of radiotherapy is conventionally given over 6 to 7 weeks using a single fraction of 1.80-2.00 Gy per day, on 5 treatment days per week. This schedule was largely developed empirically, with the 24-h interval between fractions being selected principally for patient and staff convenience. There is little radiobiological evidence to suggest that such timing is optimal in terms of the therapeutic ratio. Proliferation of clonogenic tumour cells may be significant during a 6 to 7-week course of radiotherapy and could lead to ultimate failure in some cases. Animal studies have shown that the cell kinetics of a tumour alter after irradiation, with a decrease in the potential doubling time and a decrease in the cell loss factor [2,28]. Various workers have calculated the clonogenic cell number doubling time (TclO,) in human tumours during a course of radiotherapy by comparison of the isoeffective dose for tumour control with different overall treatment times [ 17,21,22]. They found that Tclon approximated much more closely to the potential doubling time before treatment, rather than the volume doubling time before treatment. T C,onfor tumours of the head and neck region lies between 4 and 7 days. Accelerated fractionation radiotherapy (AFRT), whereby the overall treatment time is reduced by giving 2 or 3 conventional sized fractions on each treatment day, should therefore decrease the number of clonogenic cells that must be killed by the radiation [ 8,311. Shortening of the overall treatment time can be expected to increase the acute radiation reaction, as the population of clonogenic normal epithelial cells will be more drastically reduced in the short term. In contrast, late radiation effects should not be significantly worse, as these are largely dependent on the fraction size and total dose, providing there is an adequate time interval between fractions for repair of sub-lethal damage in normal tissues [ 241. The patients in this study were treated with a schedule comprising 3 fractions of 1.80 Gy, 4 h apart, on three treatment days per week (Monday,
Wednesday, Friday), and the tumour dose was taken to 59.40 Gy in 24-25 days. This regimen is a modification of the one described by Gonzales et al. [lo], who gave 3 fractions of 1.80 Gy on 5 days per vweek, and a tumour dose of 48.60-54.00 Gy in 11-12 days. When this regimen was tried in Wellington the acute reactions were felt to be unacceptably severe. In addition, the tumour had little time to regress during treatment, which might not have allowed for reoxygenation. Our modification enabled both the overall time and the tumour dose to be increased.
Patients and methods Between 1981 and 1986, 89 patients with advanced head and neck squamous cell carcinoma localised above the clavicles were treated with AFRT at Wellington Hospital. All patients gave their informed consent according to the standards of the institution. The early patients in this series were part of a pilot study which has already been reported [ 111. Sixty-nine patients received radiotherapy as primary treatment for their malignancy. Most of these patients were considered unsuitable for surgery because of advanced tumour, other significant medical conditions or old age. Twenty patients received radiotherapy for recurrent tumour following initial treatment with radical surgery. There were almost three times as many male as female patients (66 : 23). The median age of all patients was 62 years with a range of 12 to 85 years. Twenty-eight of the 69 previously untreated patients were aged 70 years or more. Site of tumour is shown in Table I. The TNM classification (UICC) of the 69 previously untreated patients is shown in Table II. The Stage Grouping of previously untreated patients with tumours of the oral cavity, pharynx and larynx is shown in Table III. Prior to treatment, patients underwent examination under anaesthetic by a head and neck surgeon, with a radiation oncologist in attend-
109 TABLE
TABLE
I
Stage grouping of previously untreated
Site of tumour. Number recurrent
Number previously untreated
5
Oral cavity Oropharynx Hypopharynx Nasopharynx Larynx glottic supraglottic Nasal cavity Skin Unknown primary Lip Parotid Total
TABLE
III
31 9 6 4
1 2 2
I
2 1
11 3 12 3 2 1
20
89
I -
69
All patients
II
TNM classification
of 69 previously untreated
NO
N,
T, T, TX T‘l TX”
1 9 10 14 3
2 3 2 10
Total
31
21
N2
TO
All patients M,. a No T classification
patients.
N3
Total
1 1 2
3 1 2 2 3
3 4 15 15 29 3
4
11
69
for nasal cavity.
ante. The diagnosis of squamous cell or undifferentiated carcinoma was confirmed histologically in all cases. Haematological and biochemical screening and chest X-ray were mandatory, and the majority of patients also had a CT scan of the neck and involved region. Dental assessment was routinely performed. All patients were treated on a Varian 4 MeV linear accelerator. Using a schedule comprising 3 fractions of 1.80 Gy, 4 h apart, on three treatment days per week (Monday, Wednesday,
Site
patients. Total
Stage I
II
III
IV
Oral cavity Oropharynx Hypopharynx Nasopharynx Larynx glottic Larynx supraglottic
1 -
2 1 -
6 3 2
18 5 4 2
26 9 5 4
1
4
._
5
5
2
2
9
Total
1
9
17
31
58
Note: There is no stage grouping for skin, nasal cavity, or unknown primary.
Friday), the tumour dose was taken to 59.40 Gy in 33 fractions in 24-25 days. Patients were treated using a shrinking field technique. Radiotherapy was initially to a large volume which included the regional lymph nodes as well as the primary tumour with a margin. Parallel opposed lateral fields were used to treat the primary tumour and upper neck, and a direct anterior field (with mid-line shielding) the lower neck. The tumour dose was taken to 37.80 Gy in 21 fractions in 15 days, with the tumour dose defined as the mid-line dose on the central axis for the parallel opposed lateral fields and the dose at 20 mm depth in the supraclavicular fossae for the direct anterior field. The minimum tumour dose was not permitted to fall below 90% of the maximum dose for the parallel opposed lateral fields. Tissue compensators were not used. The dose for subclinical disease was kept to 37.80 Gy because of uncertainty about the long term tolerance of normal tissues, particularly the spinal cord, to accelerated fractionation. Frail elderly patients did not receive prophylactic irradiation of the whole neck, as the risk of recurrence from untreated microscopic metastatic disease was outweighed by the possibility of increased morbidity from a larger treatment volume.
110
After 37.80 Gy the treatment volume was reduced to include only detectable tumour with a margin, and the spinal cord was excluded. A further 21.60 Gy in 12 fractions in 8 days was then delivered to the tumour. The fields used were either parallel opposed laterals or a technique using antero- and postero-laterals on the ipsilateral side and a direct lateral on the contralateral side. The tumour dose for the small volume was taken as the maximum dose, and the minimum tumour dose was not permitted to fall below 93% of the maximum dose. anaesthetic All patients used antiseptic mouthwash and candida prophylaxis on a regular basis. Patients who developed significant side effects were admitted to hospital to ensure that symptoms were controlled satisfactorily and that adequate nutrition was maintained. Severe mucositis was not considered a reason for suspending radiotherapy. Early in the study, 12 patients were treated with chemotherapy prior to radiotherapy. All received two or three courses of &-platinum, vinblastine and bleomycin, except for one patient with very bulky neck nodes from an unknown primary who was given high dose melphalan followed by autologous marrow rescue. Nine patients had a partial response, but there were no complete responders. Of the nine previously untreated patients who were given chemotherapy, there were six partial responders. Sixteen of the 69 previously untreated patients underwent surgery after radiotherapy for persistent tumour, relapse or radiation complications. Follow-up was calculated from the start of radiotherapy in all cases, and life-table analysis of data was performed according to the method described by Peto [25]. Complete response (CR) was defined as the absence of any clinical evidence of cancer 3 months after start of radiotherapy. Multivariate analysis of data was performed using the step-wise proportional-hazards model.
Results The median follow up of the 89 patients was 44 months, with a range of 14 to 83 months, Local-regional
control
This is defined as the control of cancer at the primary site and within the cervical lymph nodes. Figure 1 shows that for all patients, the probability of local-regional control was 39% at 3 years. However, local-regional control for the 69 previously untreated patients was 47% at 3 years, whereas for the 20 patients treated for recurrence after radical surgery local-regional control was only 12% at one year (p < 0.001). The CR rate in the 69 previously untreated patients was 68%. In this group there were 33 local-regional failures - one case of nodal relapse outside the radiation field, two cases of nodal relapse within the prophylactic field and 30 cases of persistent or recurrent tumour at sites of presenting disease. Figure 2 shows the probability of local-regional control according to stage for 58 previously untreated patients with tumours of the oral cavity, pharynx or larynx. The whole group had a CR rate of 7 1 y0 and a 3 year local regional control rate of
Fig. 1. Probability of local-regional control previous treatment status.
according
to
111 Distant metastases Of the 69 previously untreated patients, nine developed distant metastases, but only two of these did not have associated local-regional failure. Survival
Fig. 2. Probability of local-regional control according to stage for patients with previously untreated tumours of oral cavity, pharynx or larynx.
49%. There was a 100% CR rate for the 27 patients with Stage I, II or III tumours, whereas the CR rate for the 31 patients with Stage IV tumours was only 45 % . The probability of local-regional control was strongly dependent upon stage (p < O.OOOl), and at 3 years was loo%, 78% and 15% for Stage I + II, Stage III and Stage IV tumours, respectively. Table IV shows the results of multivariate analysis of risk factors for local-regional failure in the 58 previously untreated p.atients with tumours of the oral cavity, pharynx or larynx. Only the T and N staging reached statistical significance, whereas age, sex, site and grade of tumour did not.
TABLE
There were 56 deaths in the 89 patients treated, of which 50 were due to cancer. The remaining six patients died of intercurrent disease (three patients of myocardial infarction, one of a cerebra-vascular accident, one of a gastrointestinal haemorrhage and one of a motor vehicle accident) and were free of clinical evidence of cancer at the time of death. Figure 3 shows that the probability of survival from cancer for the 69 previously untreated patients was 44% at 3 years, whereas the probability of survival for the 20 patients treated for recurrence after surgery was only 20% at 2 years (p < 0.01). Figure 4 shows the probability of survival from cancer according to stage for 58 previously untreated patients with tumours arising in the oral cavity, pharynx or larynx. The probability of survival was strongly dependent on stage
PRO$ABLlTY
IV
Multivariate
analysis of risk factors for local-regional failure.
Factor
P
T,T,T3T,
< 0.000 1
&N&N, Site: Oral cavity/pharynx/larynx Grade: l/2/3,4 Age (yrs): < 65/> 65 Sex: M/F
0.03 0.73 0.68 0.25 0.12
Fig. 3. Probability
of survival according ment status.
to previous treat-
dysphagia due to the mucositis was strongly dependent on the site of the primary tumour. Patients with tumours of the oral cavity or pharynx often required admission during the third week of treatment to ensure that adequate nutrition was maintained. In contrast, patients with tumours of the larynx and other primary sites were usually able to complete treatment as outpatients. In all, 46 of the 89 patients (52%) required admission, and the mean duration of admission was 18 days. 28 patients (31%) required treatment with total parenteral nutrition (TPN) because oral intake was inadequate and their weight continued to fall despite opiates and a variety of topical treatments. The mean duration of TPN when used was 12 days. Skin reactions were much less severe than the mucosal reactions, probably due to the skin sparing effect of the megavoltage radiation. When the skin sparing effect was lost due to the surface contour, localised moist desquamation usually developed by the end of treatment to the high dose volume (week 4). Otherwise, moderate to marked erythema with some dry desquamation was the standard maximal reaction. A degree of xerostomia occurred in all patients, and developed during the second week of treatment. However, it only became severe if both parotid salivary glands were included in the high dose volume, and then exacerbated the dysphagia due to mucositis. Acute radiation reactions were slightly exaggerated in 11 of the 12 patients who received prior
Fig. 4. Probability of survival according to stage for patients with previously untreated tumours of oral cavity, pharynx or larynx.
(p < 0.0001) and at 3 years was 90x, 75% and 15% for Stage I and II, Stage III and Stage IV tumours, respectively. The 12 patients treated with initial chemotherapy had a similar survival to patients treated with radiotherapy alone. Complications Acute The acute mucosal reactions were generally severe. Table V shows the mean mucosal reaction according to time and accumulated dose for 21 recently treated patients who had mucosa included in the high dose volume. The severity of
TABLE
V
Mean mucosal reaction according to accumulated
dose and time.
Week Accumulated dose (Gy) Mucosal reaction (mean score)a
0
1
2
3
4
0
16.2
32.4
48.6
59.4
0
1.1
2.8
4.1
4.1
a Modified from Wang [30]. 0, Normal appearance; 1, mild erythema; mucositis.
2, severe erythema;
3, studded
mucositis;
5
6
8
10
3.4
2.6
1.4
0.6
4, confluent
mucositis;
5, haemorrhagic
113 chemotherapy. However, in one patient who was initially treated with high dose melphalan and autologous marrow rescue, the reaction was markedly exaggerated. This patient with neck nodes from an unknown primary developed prolonged grade 5 mucositis and extensive moist desquamation of the skin. He subsequently developed severe late complications. There were three deaths attributed to acute toxicity in the 89 patients, with two cases of bronchopneumonia and one of septicaemia. For statistical purposes, these were considered as cancer deaths. All toxic deaths occurred in patients with recurrent or Stage IV tumours. One patient had received prior treatment with chemotherapy. Late In this report, late complications are those persisting or developing 3 months or more after completion of radiotherapy.
Other late complications. One patient with a T,, N, carcinoma of the supraglottic larynx required total laryngectomy for radionecrosis 7 months after completion of radiotherapy. Interestingly, at the time of surgery he was found to be hypothyroid. Another patient with a T,, N, carcinoma of the oral cavity has moderately severe trismus 38 months after treatment. Salvage surgery Sixteen of the 69 previously untreated patients underwent surgery following radiotherapy. Two patients required surgery for radionecrosis, and three had excision of suspicious neck nodes which histologically showed no cancer. The other 11 patients had limited excision of persistent or recurrent cancer and two of these have since remained free of cancer.
Discussion Mucosa. Asymptomatic slight to moderate mucosal oedema persisted in 14 patients for up to 8 months after radiotherapy. It was more often a feature in advanced tumours of the larynx. Slight to moderate xerostomia has persisted in 18 patients with oropharyngeal or oral cavity tumours when one or both parotid salivary glands were included in the high dose volume. Skin and subcutaneous tissues. One patient developed localised skin necrosis 22 months after radiotherapy, which necessitated full thickness skin grafting of the affected area. This patient with neck nodes from an unknown primary was treated with high dose melphalan and autologous marrow rescue prior to radiotherapy, and also experienced exaggerated acute radiation effects. Two other patients developed asymptomatic subcutaneous fibrosis in the anterior neck. In both, a marked variation in surface contour resulted in dosimetry problems which might have been avoided by using tissue compensators. Satisfactory healing of the skin and other tissues occurred in all 16 patients who underwent surgery after radiotherapy.
As expected, the acute complications of our regimen of AFRT were considerably more severe than those of conventional fractionation, with mucositis a particular problem. However, we decided against modtication of our treatment regimen and opted instead for a policy of intensive support to tide the patient over the acute reaction. This necessitated the admission of 52% of patients, compared to our previous experience of 2 1% for conventionally fractionated radiotherapy (66.00 Gy in 33 fractions over 45 days). Patient management became easier as experience of our AFRT regimen increased and the timing of mucosal reactions could be accurately predicted. The three toxic deaths all occurred in the early part of the study, and although these patients all had very advanced tumour, it is possible their deaths might have been avoided if their problems had been anticipated. Later our policy was to readily admit patients with extensive tumours of the oral cavity and pharynx when the mucosal reaction became severe, and to initiate parenteral feeding as soon as oral intake became inadequate.
114 This approach permitted treatment of even quite elderly patients, and 28 of the previously untreated patients were aged 70 years or more. The late complications of our AFRT regimen have been quite acceptable, and not obviously worse than our previous experience with conventional fractionation. Of the 69 previously untreated patients, only three developed moderate to severe late complications, with one case each of laryngeal necrosis, skin necrosis and trismus. Our incidence of laryngeal necrosis is 7 % (one of 14 patients), identical to the 5-12x incidence reported in the literature for conventional fractionation [ 71. Healing was not a problem in those patients who underwent surgery after AFRT. Our results support the widely held belief that any treatment used for recurrence is much less likely to be effective than the same treatment given ab initio. Also, the probability of local-regional control and survival are similar at any given time, confirming that survival is largely dependent on achieving local-regional control [ 181. Our 3 year local-regional control rate of 47%
TABLE
for all previously untreated patients is comparable to recently published results for patients treated with conventionally fractionated radiotherapy, where local-regional control rates of 26-&l% were achieved [ 4,6,29]. Our previously untreated patients with tumours of the upper aero-digestive tract were analysed in greater detail as they comprise a less heterogeneous population compared to the group as a whole. We used the Stage Grouping System in order to obtain sub-groups with sufhcient numbers for statistical comparison. The 3 year localregional control rate was 78% for 17 Stage III tumours and 15% for 31 Stage IV tumours. The figure for Stage III tumours is encouraging, but there are to few patients and follow-up is too short for any firm conclusions. Table VI summarises other reported studies of AFRT for head and neck cancer. It should be noted that many studies are not pure AFRT, as either a reduction has been made in the total dose, or there has been no appreciable reduction in the overall treatment time because a split has been
VI
AFRT for head and neck cancer. Author
No. of patients
Dose/F (Gy)
No. of fractions per day
Total dose (Gy)
Time (wks)
Split (wks)
Results
CR at 8 months 68% CR at 3 months 86% 3-Year survival 44% 2-Year actuarial local control rate 65% Highly significant increase in 3-year actuarial local control rate (68% vs.
Perracchia and Salti [23] Svoboda [27]
22 59
2 1.75-2.3
3 3
48-54 50-55
-1.5 -2
-
Knee et al. [ 161
53
1.2-2.0
2
- 70-74
6
-
321
1.6
2
64-67.2
6
2
303 a 308
1.8 1.6
1 3
-65 67.2-72
7.2 6-7
3-4
159b
2
1
7
-
Wang [30]
46%) Van den Bogaert and van der Schueren [29]
a Historical controls. ’ Randomised controls.
70
No difference in 3-year actuarial local control rate (35% for all patients)
115 introduced [ 241. These compromises have been made-in an attempt to reduce the acute normal tissue reactions. The only randomised study comparing AFRT to conventional fractionation has been performed by the EORTC [29], and this showed no significant difference in the 3 year local control rate, which was 35 % for all patients. However, the overall treatment time for AFRT with a split and conventional fractionation was almost identical. Wang [ 301 also employed a split in his schedule, but still reduced the overall treatment time by over a week. His 3 year localregional control rate for 321 patients was 61 y0 compared to 46% for historical controls. Knee and colleagues [ 161 at the M.D. Anderson Hospital used the concomitant boost technique, a variation of AFRT whereby the boost is delivered as a second daily fraction during the basic treatment course. Their regimen reduced the total duration of treatment by about 2 weeks. They treated a heterogeneous group of 53 patients with a tumour dose of 70-74 Gy. The local-regional control rate at 2 years was 65 %, but there was a relatively high incidence of moderate to severe late complications. Our continuous AFRT schedule is feasible and merits further investigation. There is now enough accumulated experience in AFRT for advanced head and neck cancer to justify a multi-centre randomised study comparing a continuous schedule of AFRT with conventional fractionation.
Acknowledgements The authors would like to thank Dr. P. J. Dady, Mr. J. Hornibrook, Mr. P. Blake, Mr. N. Thomson, Mr. D. Palmer and Mr. J. Scott, for their help in the management of these patients. They would also like to thank Miss S. Parker for typing the manuscript and Dr. P. B. Roberts for his advice.
References 1 Berry, G. H., Dixon, B. and Ward, A. J. The Leeds results for radiotherapy in H.B.O. for carcinoma of the head and neck. Clin. Radiol. 30: 591-592, 1979. 2 Denekamp, J. Cell Kinetics and Cancer Therapy. C. C. Thomas, Springfield, IL, 1982. 3 Denham, J. W., Yeoh, E. K., Wittwer, G., Ward, G. G., Ahmad, A. S. and Harvey, N. D. M. Radiation therapy in hyperbaric oxygen for head and neck cancer at Royal Adelaide Hospital - 1964 to 1980. Int. J. Radiat. Oncol. Biol. Phys. 13: 201-208, 1987. 4 Duncan, W., Orr, J. A., Arnott, S. J., Jack, W. J. L., Kerr, G. R. and Williams, J. R. Fast neutron therapy for squamous cell carcinoma in the head and neck region: results of an randomised trial. Int. J. Radiat. Oncol. Biol. Phys. 13: 171-178, 1987. 5 Fazekas, J. T., Sommer, C. and Kramer, S. Adjuvant intravenous methotrexate or definitive radiotherapy alone for advanced squamous cancers of the oral cavity, oropharynx, supraglottic larynx or hypopharynx: concluding report of an RTOG randomised trial on 638 patients. Int. J. Radiat. Oncol. Biol. Phys. 6: 533-541, 1980. 6 Fazekas, J., Pajak, T. F., Wasserman, T., Marcia& V., Davis, L., Kramer, S., Rotman, M. and Stetz, J. Failure of misonidazole sensitised radiotherapy to impact upon outcome among stage III-IV squamous cancers of the head and neck. Int. J. Radiat. Oncol. Biol. Phys. 13: 1155-l 160, 1987. I Ferguson, B. J., Hudson, W. R. and Farmer, J. C. Hyperbaric oxygen therapy for laryngeal radionecrosis. Ann. Otol. Rhinol. Laryngol. 96: l-6, 1987. 8 Fowler, J. F. Potential for increasing the differential response between tumors and normal tissues: can proliferation rate be used? Int. J. Radiat. Oncol. Biol. Phys. 12: 641-645, 1986. 9 Fu, K. K., Phillips, T. L., Silverberg, I. J., Jacobs, C., Goffinet, D. R., Chun, C., Friedman, M. A., Kohler, M., McWhirter, K. and Carter, S. K. Combined radiotherapy and chemotherapy with bleomycin and methotrexate for advanced inoperable head and neck cancer: update of a Northern California Oncology Group randomised trial. J. Clin. Oncol. 5: 1410-1418, 1987. 10 Gonzales, D., Breur, K. and van der Schueren, E. Preliminary results in advanced head and neck cancer with radiotherapy by multiple fractions a day. Clin. Radio]. 3 1: 417-421, 1980. 11 Gray, A. J. Treatment of advanced head and neck cancer with accelerated fractionation. Int. J. Radiat. Oncol. Biol. Phys. 12: 9-12, 1986. 12 Haselow, R. E., Adams, G. S., Oken, M. M., Gondsmit, A., Lerner, H. J. and Marsh, J. C. Cis-platinum (DDP) with radiation therapy (RT) for locally advanced unresectable head and neck cancer. Proc. Am. Sot. Clin. Oncol. 2: 160, 1983.
116 13 Henk, J. M., Kunkler, P. B. and Smith, C. W. Radiotherapy and hyperbaric oxygen in head and neck cancer. Lancet II: 101-103, 1977. 14 Henk, J. M. and Smith, C. W. Radiotherapy and hyperbaric oxygen in head and neck cancer. Interim report of 2nd clinical trial. Lancet II: 104-105, 1977. 15 Kaplan, M. J., Hahn, S. S., Johns, M. E., Stewart, F. M., Constable, W. C. and Cantrell, R. Mitomycin and fluorouracil with concomitant radiotherapy in head and neck cancer. Arch. Otolaryngol. 111: 220-222, 1985. 16 Knee, R., Fields, R. S. and Peters, L. J. Concomitant boost radiotherapy for advanced squamous cell cancer of the head and neck. Radiother. Oncol. 4: l-7, 1985. 17 Maciejewski, B., Gunhild, P-B. and Trott, K-R. The influence of the number of fractions and of overall treatment time on local control and late complication rate in squamous cell carcinoma of larynx. Int. J. Radiat. Oncol. Biol. Phys. 9: 321-328, 1983. 18 Merino, 0. R., Lindberg, R. D. and Fletcher, G. H. An analysis of distant metastases from squamous cell carcinoma of the upper respiratory and digestive tracts. Cancer 40: 145-151, 1977. 19 Million, R. R. and Cassisi, N. J. Management of Head and Neck Cancer. J.B. Lippincott Company, Philadelphia, 1984. 20 Murthy, A. K., Taylor, S. G., Showel, J., Caldarelli, D. D., Hutchinson, J. C., Holinger, L. D., Witt, T. R. and Hoover, S. Treatment of advanced head and neck cancer with concomitant radiation and chemotherapy. Int. J. Radiat. Oncol. Biol. Phys. 13: 1807-1813, 1987. 21 Niederer, J., Hawkins, N. V., Rider, W. D. and Till, J. E. Failure analysis of radical radiation therapy of supraglottic laryngeal carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 2: 621-629, 1976. 22 Parsons, J., Bova, F. J. and Million, R. R. A re-evaluation of split-course technique for squamous cell carcinoma of the head and neck. Int. J. Radiat. Oncol. Biol. Phys. 6: 1645-1652, 1980.
23 Peracchia, G. and Salti, C. Radiotherapy with thrice-aday fractionation in a short overall time. Clinical experiences. Int. J. Radiat. Oncol. Biol. Phys. 7: 99-104, 1981. 24 Peters, L. J. and Ang, K. K. Unconventional fractionation schemes in radiotherapy. In: Important Advances in Oncology, pp. 269-286. Editors: V. J. De Vita, S. Hellman and S. A. Rosenberg. J.B. Lippincott Company, Philadelphia, 1986. 25 Peto, R., Pike, M. C., Armitage, P., Breslow, N. E., Cox, D. R., Howard, S. V., Mantel, N., McPherson, K., Peto, J. and Smith, P. G. Design and analysis of randomised clinical trials requiring prolonged observation of each patient. II. Analysis and examples. Br. J. Cancer 35: l-39, 1977. 26 Seely, R. and Cridland, S. The treatment of locally advanced head and neck cancer with misonidazole, hyperbaric oxygen and irradiation. Int. J. Radiat. Oncol. Biol. Phys. 10: 1721-1723, 1984. V. H. J. Accelerated fractionation: the 27 Svoboda, Portsmouth experience 1972-1984. In: Proceedings of Varian’s Fourth European Clinac Users Meeting, Malta, May 25-26, pp. 70-75. Varian, Zug, Switzerland, 1984. 28 Trott, K-R. and Kummermehr, J. What is known about tumour proliferation rates to chose between accelerated fractionation or hyperfractionation? Radiother. Oncol. 3: l-9, 1985. 29 Van den Bogaert, W. and van der Schueren, E. Early results of the EORTC randomised clinical trial on multiple fractions per day (MFD) and misonidazole in advanced head and neck cancer. Int. J. Radiat. Oncol. Biol. Phys. 12: 587-591, 1986. 30 Wang, C. C., Blitzer, P. H. and Suit, H. Twice a-day radiation therapy for cancer of the head and neck. Cancer 55: 2100-2104, 1985. 31 Withers, H. R. Biologic basis for altered fractionation schemes. Cancer 55: 2086-2095, 1985.
