Fast neutron radiation therapy for unresectable squamous cell carcinomas of the head and neck: the results of a randomized RTOG study

Interventions for the treatment of oral cavity and oropharyngeal cancer: radiotherapy

BACKGROUND

The management of advanced oral cavity and oropharyngeal cancers is problematic and has traditionally relied on surgery and radiotherapy, both of which are associated with substantial adverse effects. Radiotherapy has been in use since the 1950s and has traditionally been given as single daily doses. This method of dividing up the total dose, or fractionation, has been modified over the years and a variety of approaches have been developed with the aim of improving survival whilst maintaining acceptable toxicity.

OBJECTIVES

To determine which radiotherapy regimens for oral cavity and oropharyngeal cancers result in increased overall survival, disease free survival, progression free survival and locoregional control.

SEARCH STRATEGY

The following electronic databases were searched: the Cochrane Oral Health Group's Trials Register (to 28 July 2010), CENTRAL (The Cochrane Library 2010, Issue 3), MEDLINE via OVID (1950 to 28 July 2010) and EMBASE via OVID (1980 to 28 July 2010). There were no restrictions regarding language or date of publication.

SELECTION CRITERIA

Randomised controlled trials where more than 50% of participants had primary tumours of the oral cavity or oropharynx, and which compared two or more radiotherapy regimens, radiotherapy versus other treatment modality, or the addition of radiotherapy to other treatment modalities.

DATA COLLECTION AND ANALYSIS

Data extraction and assessment of risk of bias was undertaken independently by two or more authors. Study authors were contacted for additional information as required. Adverse events data were collected from published trials.

MAIN RESULTS

30 trials involving 6535 participants were included. Seventeen trials compared some form of altered fractionation (hyperfractionation/accelerated) radiotherapy with conventional radiotherapy; three trials compared different altered fractionation regimens; one trial compared timing of radiotherapy, five trials evaluated neutron therapy and four trials evaluated the addition of pre-operative radiotherapy. Pooling trials of any altered fractionation radiotherapy compared to a conventional schedule showed a statistically significant reduction in total mortality (hazard ratio (HR) 0.86, 95% confidence interval (CI) 0.76 to 0.98). In addition, a statistically significant difference in favour of the altered fractionation was shown for the outcome of locoregional control (HR 0.79, 95% CI 0.70 to 0.89). No statistically significant difference was shown for disease free survival.No statistically significant difference was shown for any other comparison.

AUTHORS' CONCLUSIONS

Altered fractionation radiotherapy is associated with an improvement in overall survival and locoregional control in patients with oral cavity and oropharyngeal cancers. More accurate methods of reporting adverse events are needed in order to truly assess the clinical performance of different radiotherapy regimens.

Particle beam radiotherapy for head and neck tumors: radiobiological basis and clinical experience

BACKGROUND

Head and neck tumors are often located near critical organs, making it impossible to deliver a dose of conventional radiotherapy high enough to eradicate the disease. Our aim was to review the potential benefits and available clinical experience of particle beam therapy (hadrontherapy) in the treatment of these tumors.

METHODS

A review of the literature was carried out through a MEDLINE search (publications between 1980 and 2005).

RESULTS

A review of the available clinical data shows that particle beam therapy can offer several radiobiological and physical advantages over conventional photon radiotherapy: improved dose distribution permits dose escalation within the target and optimal sparing of normal tissue. Preclinical and clinical studies suggest that there may be benefits to using hadrontherapy for tumors characterized by poor radiosensitivity and critical location. At present, the most used hadrons are protons and, as yet on an experimental basis, carbon ions. It is now well accepted that there are certain indications for using proton therapy for skull base tumors (chordoma and chondrosarcoma), paranasal sinus carcinomas, selected nasopharyngeal tumors, and neutron/ion therapy for salivary gland carcinomas (in particular, adenoid cystic tumors). Its viability in other cases, such as locally advanced squamous cell carcinoma, melanoma, soft tissue sarcoma, and bone sarcoma, is still under investigation.

CONCLUSIONS

Hadrontherapy can be beneficial in the treatment of tumors characterized by poor radiosensitivity and critical location. Further clinical and radiobiological studies are warranted for improved selection of patient population.

The use of unequal randomisation ratios in clinical trials: a review

OBJECTIVE

To examine reasons given for the use of unequal randomisation in randomised controlled trials (RCTs).

MAIN MEASURES

Setting of the trial; intervention being tested; randomisation ratio; sample size calculation; reason given for randomisation.

METHODS

Review of trials using unequal randomisation. DATABASES AND SOURCES: Cochrane library, Medline, Pub Med and Science Citation Index.

RESULTS

A total of 65 trials were identified; 56 were two-armed trials and nine trials had more than two arms. Of the two-arm trials, 50 trials recruited patients in favour of the experimental group. Various reasons for the use of unequal randomisation were given. Six studies stated that they used unequal randomisation to reduce the cost of the trial, with one screening trial limited by the availability of the intervention. Other reasons for using unequal allocation were: avoiding loss of power from drop-out or cross-over, ethics and the gaining of additional information on the treatment. Thirty seven trials papers (57%) did not state why they had used unequal randomisation and only 14 trials (22%) appeared to have taken the unequal randomisation into account in their sample size calculation.

CONCLUSION

Although unequal randomisation offers a number of advantages to trials the method is rarely used and is especially under-utilised to reduce trial costs. Unequal randomisation should be considered more in trial design especially where there are large differences between treatment costs.

Dose escalation study of carbon ion radiotherapy for locally advanced head-and-neck cancer

PURPOSE

To evaluate the toxicity and efficacy of carbon ion radiotherapy for head-and-neck cancer in a Phase I/II dose escalation clinical trial.

METHODS AND MATERIALS

Between June 1994 and January 1997, 36 patients with locally advanced, histologically proven, and new or recurrent cancer of the head and neck were treated with carbon ions. A dose escalation study was conducted, delivering 18 fractions through 6 weeks for 17 patients (Group A) and 16 fractions through 4 weeks for 19 patients (Group B). Eligibility and ineligibility criteria were the same in both groups. The dosages were escalated in increments of 10% after careful observation of at least 3 patients treated with the same dosages. The endpoints of the study were a Grade 3 reaction of the skin and the mucous membrane or local control of the tumors.

RESULTS

Follow-up time ranged from 77 to 108 months with a median of 90 months. Grade 3 acute reaction of the skin was detected in 1 of the 2 patients in Group A who were treated with 70.2 GyE/18 fractions/6 weeks. In Group B, Grade 3 acute skin reaction was detected in 20% (1/5), 27% (2/11), and 67% (2/3) patients treated with 52.8 GyE, 57.6 GyE, and 64.0 GyE through 16 fractions for 4 weeks, respectively. There was only 1 patient with a Grade 3 acute reaction of the mucous membrane. Only 1 patient developed a Grade 2 late reaction of the mucous membrane (superficial ulcer), which was located close to the tumor. No other Grade 2 or greater late reaction was noted until the time of analysis. Acute tumor reactions in 34 patients consisted of 10 patients of complete response 19 of partial response, 4 of no change, and 1 of progressive disease. Local control of 34 patients calculated by the Kaplan-Meier method was 75% at 5 years. Five years' local control of five malignant melanomas showed 100%, and that of 9 patients with adenoid cystic carcinoma was 90%. Also, local control of 8 patients of salivary glands and 4 patients of ears was 100% at 56 months and 5 years.

CONCLUSIONS

The dose fractionation methods of 70.2 GyE through 18 fractions for 6 weeks and 64.0 GyE through 16 fractions for 4 weeks showed equal clinical outcome in terms of morbidity and local control. The outcome of carbon ion radiotherapy showed a specific effectiveness in local control of non-squamous cell carcinoma such as adenoid cystic carcinomas and malignant melanomas. From the results of this study, it can be concluded that carbon ion radiotherapy will deliver a high local control rate without unacceptable injuries to the surrounding normal tissues.

Le traitement par neutrons : hadronthérapie partie II : bases physiques et expérience clinique

Neutrons have radiobiological characteristics, which differ from those of conventional radiotherapy beams (photons) and which offer a theoretical advantage over photons to fight radioresistance by the differential relative biological effect of them between normal and tumour tissues. Neutron therapy beneficed of great interest between 1975 and 1985. Many of phase III trials were conducted and indications have been definitively deducted of them. After briefly describing the properties of neutron beams, this review discusses the indication of neutron therapy on the basis of the clinical results. Salivary, prostate tumours and sarcomas are the main indications of neutron therapy. In concern to the prostate cancers, other alternative treatments reduce the neutron therapy field. For sarcomas, the lack of randomised trials limits the impact of the interest of neutrons. For other tumours, the ratio benefice/risk of neutron therapy is inferior to these obtained with photons and they could not be considered like classical indications.

Monte Carlo simulation of fast neutron spectra: mean lineal energy estimation with an effectiveness function and correlation to RBE

PURPOSE

Intercomparisons of radiotherapy trials conducted at different fast neutron facilities are complicated by the dependence of the relative biologic effectiveness (RBE) of the different beams on the fast neutrons spectra. To obtain a better understanding of the influence of neutron energy on radiation quality, Monte Carlo simulations were performed to calculate fast neutron (FN) spectra at different irradiation positions. To allow for comparisons with experimental data, the positions were chosen to be the same as that used by other investigators to obtain microdosimetry readings and radiobiological data.

METHODS AND MATERIALS

The primary neutron yield for beryllium targets bombarded with protons at the National Accelerator Center, Louvain, Nice, and Orleans facilities were calculated using the FLUKA code. Neutron transport simulations were performed with MCNP-4A, giving FN spectra for various phantom depths, hardening filter thickness, and field sizes. Using an effectiveness function, FN energy groups were correlated with mean lineal energies (y*-values) obtained experimentally by other workers.

RESULTS

Calculations confirm earlier measurements that a decrease in beam quality by a hardening filter is the result of a reduction in the low-energy neutron component, i.e., neutrons below 3 MeV. Variations in RBE due to changes in field size and different phantom depths could also be explained by variations of neutrons with energies between 3-15 MeV. The effectiveness function allows one to calculate changes in y* observed for the NAC beam with great accuracy (R(2) = 0.99, p < 0.0001). Also, when this function is applied to beams with different neutron energies, y* calculated values show a very significant correlation with measured RBE values (R(2) = 0.98, p < 0.0001).

CONCLUSION

The effectiveness function appears to be suitable to predict changes in y*-values and variations in RBE, using FN spectra simulated for various neutron therapy facilities.

Fast Neutron Irradiation in Advanced Pre-Irradiated Head and Neck Tumors

Aims and background

Radiotherapy with fast neutrons offers radiobiological advantages in the treatment of hypoxic and slowly proliferating tumors. Tumors recurring in pre-irradiated areas of the head and neck usually exhibit a low radiation sensitivity that seems to promise little success for a repeated irradiation with photons. In such a situation, radiotherapy with fast neutrons may offer an alternative treatment option. To evaluate the question, the Muenster patients were retrospectively analyzed.

Methods

During the period July 1986 to January 1994, 26 patients with local-regional recurrences of pre-irradiated head and neck tumors were treated with fast neutrons at our department.

Results

Median survival calculated by the Kaplan-Meier method was 7.4 months. The 1-year survival probability was 29.3% and the 2-year survival probability 5.9%. The objective tumor remission (complete + partial) was 50%. A subjective improvement of symptoms was reported by 42.3% of all patients.

Conclusions

Despite relatively unfavorable survival rates produced by the therapy, good remission rates and thereby a good palliative effect can be attained with a short treatment time and tolerable side effects.

The role of fast neutrons in the treatment of squamous cell carcinomas of the head and neck: The European experience

After the first European stet on neutron therapy of head and neck cancer, published by Catterall and Bewley, revealed significant advantages of neutrons over photons in 134 randomized patients, the subsequent phase II and phase III studies led to controversial results. Figures for local tumor control varied from 26-76% after 2 years to 19% after 10 years and for survival from 16-46% after 2 years to 14% after 10 years. Results of all studies were consistent with showing an increased incidence of severe late effects. Eligibility criteria with regard to the tumor stage, history, and previous therapies of relapses varied from study to study. Also, physical treatment parameters, such as neutron energy, LET and OER, neutron and photon doses for one fraction and for the total radiotherapy were rather different. The benefit of neutron therapy in comparison with photons was low or non-existent when smaller tumor stages, such as primaries T2 or lymph nodes N1 were included into the trial. It increased when only T4 tumors, recurrences or tumors persisting after conventional radiotherapy were treated and when the percentage of fixed lymph nodes increased. There was only one European study in which patients suffering form lymph node metastases of the neck had, in contrast to American data (Griffin et al. 1978), a worse prognosis after neutrons than after photon therapy (Duncan et al. 1987b). In our own phase II study on reactor neutron therapy of 100 patients suffering from relapses and persistent tumors, not yet published, 3-year survival of 21% and 3-year local control of 32% were obtained. Although this was not a randomized study, during this period all patients with equivalent tumors treated by photon therapy alone died within 2 years. In conclusion there is no general indication for neutron therapy in squamous cell carcinoma of the head and neck, but there is a benefit of neutrons in palliative treatment of recurrences, necrotic and hypoxic primaries and lymph nodes.

The merits of cell kinetic parameters for the assessment of intrinsic cellular radiosensitivity to photon and high linear energy transfer neutron irradiation

PURPOSE

Differences in tumor response and intrinsic cellular radiosensitivity make the selection of patients for specific radiation modalities very difficult. The reasons for these differences are still unclear, but are thought to be due to genomic and cellular characteristics. Because radiosensitivities vary between cell cycle stages and because S phase cells are very radioresistant, cell cycle kinetic parameters could be a candidate for predicting intrinsic radiosensitivity.

METHODS AND MATERIALS

A panel of 15 tumor cell lines was analyzed for S phase content and potential doubling times (Tpot), and the influence of these parameters on the intrinsic radiosensitivity to 60Co gamma- and p(66)/Be neutron irradiation was assessed.

RESULTS

S phase content and Tpot show a statistically significant correlation with the mean inactivation dose for photons. The correlation between cell kinetic parameters and the mean inactivation dose for neutrons showed the same trend as photon sensitivity but this was not found to be statistically significant.

CONCLUSIONS

S phase content and Tpot were identified as suitable criteria for predicting photon sensitivity. It is suggested that cell kinetic parameters could play a role in identifying neutron sensitive tumors if both tumor and normal cells are analyzed.

Clinical trials of neutron radiotherapy in the United States

The development of clinical neutron facilities in the 1980s, capable of delivering high energy neutrons spurred full scale phase III testing of neutron beam radiotherapy in a number of tumors including salivary gland, head and neck, prostate, and non small-cell lung cancer. The Radiation Therapy Oncology Group (RTOG) and the Medical Research Council (MRC) jointly sponsored a randomized trial for the treatment of advanced stage salivary gland tumors comparing neutron to conventional photon and/or electron radiotherapy. Although no improvement in survival was seen, the study demonstrated a striking and statistically significant difference in the local-regional control of unresectable salivary gland tumors (56 vs 17%), favoring neutron beam irradiation. Subsequent clinical trials of neutron beam irradiation were initiated by the Neutron Therapy Collaborative Working Group (NTCWG) sponsored by the National Cancer Institute (NCI). A phase III trial comparing neutron to photon radiotherapy for inoperable regional non-small cell lung cancer showed no overall improvement in survival. However, a statistically significant improvement in survival was observed in the subset of patients with squamous cell histology. The NTCWG trial comparing fast-neutron therapy versus conventional photon irradiation in the treatment of advanced squamous cell carcinomas of the head and neck showed a statistically significant improvement in initial complete response (70 vs 52%) favoring neutrons. However, subsequent failures erased any difference in ultimate local-regional control rates and survival curves were essentially the same in both arms. The randomized study of the NTCWG for locally advanced prostate cancer demonstrated a significant decrease in local-regional failure (11 vs 32%) at 5 years, favoring the neutron arm. Furthermore, biochemical measures of disease control also favored the neutron arm with prostate specific antigen (PSA) levels elevated in 17% of the neutron-treated patients compared to 45% of the photon-treated patients at 5 years. At the 5-year analysis, no significant difference in survival was observed between the two arms; however, longer follow-up is necessary to assess the ultimate impact of improved local-regional control on survival. An analysis of complications in this series revealed the importance of beam shaping and treatment planning capabilities in maintaining long-term sequelae following neutron irradiation at an acceptably low level.

Ultrastructural changes in the cochlea of the guinea pig after fast neutron irradiation

Guinea pigs were irradiated with fast neutrons. After a single dose of 2, 6, 10, or 15 Gy was applied, scanning and transmission electron microscopy of the temporal bone was performed to assess the effect of fast neutron irradiation on the cochlea. Outer hair cell damage appeared with neutron irradiation of more than 10 Gy, and inner hair cell damage with neutron irradiation of more than 15 Gy. Outer hair cells were more severely damaged than inner hair cells. No statistically significant differences were found in damage of basal, middle, and apical turns. The second and third rows of outer hair cells were more severely damaged than the first row of outer hair cells. The most significant findings in transmission electron microscopy were clumping of chromatin and extension of the heterochromatin in the nuclei of hair cells. The cytoplasmic changes were sequestration of cytoplasm, various changes of mitochondria, formation of vacuoles, and irregularly arranged stereocilia. The morphologic change in stria vascularis was intercellular and perivascular fluid accumulation. It appeared to be a reversible process.

Development of fast neutron therapy worldwide. Radiobiological, clinical and technical aspects

Radiobiological data indicate that fast neutrons could bring a benefit in the treatment of some tumour types, and suggest mechanisms through which this benefit could be achieved. However, radiobiology also clearly indicates that there is a need for patient selection as well as for a high-physical selectivity. The main difficulty when interpreting the results of neutron therapy are the poor technical conditions in which the first treatments were applied. This explains why the value and the place of neutron therapy are not universally recognized, although more than 15,000 patients have been treated so far worldwide. There are, however, clinical indications of fast neutrons bringing a benefit for the following tumour sites: salivary glands, paranasal sinuses, soft tissue sarcomas, prostatic adenocarcinomas, palliative treatment of melanoma and rectum. These tumours represent about 10-15% of all patients currently referred to the radiation therapy departments.

Report of long-term follow-up in a randomized trial comparing radiation therapy and radiation therapy plus hyperthermia to metastatic lymph nodes in stage IV head and neck patients

PURPOSE

The treatment of inoperable metastatic lymph nodes in patients with head and neck cancer represents a therapeutic challenge. Clinical results using conventional radiation therapy are disappointing; on the other hand, the evaluation of recent innovative radiotherapeutic methods is still pending. The end points of this analysis were focused on long-term local control, on its potential influence on survival, and on late toxicity of a previously reported randomized Phase III study comparing conventionally fractionated radical irradiation alone or combined with local hyperthermia in fixed and inoperable metastatic neck lymph nodes.

METHODS AND MATERIALS

The medical records of 41 patients (44 nodes) with advanced locoregional Stage IV squamous cell cancer of the head and neck and randomized to treatment in the period 1985-1986 with irradiation alone (22/23 evaluable nodes) or combined with external hyperthermia (18/21 evaluable nodes), were re-evaluated.

RESULTS

The statistically significant difference observed in "early" response (p = 0.0164) in favor of the combined treatment results in improved 5-year actuarial nodal control (p = 0.015). Clinical improvement noted in tumor control positively affects survival, leading to a statistically significant difference in survival at 5 years (p = 0.02). With respect to side effects, no clearly enhanced acute or late toxicity has been found; as severe late effects, two patients with bone necroses possibly related to the combined treatment have been observed. Thermal analysis failed to show a significant correlation between heating parameters and the end points of the study.

CONCLUSION

This report with 5-year follow-up confirms the efficacy and the absence of severe toxicity of the combination of radical radiation and hyperthermia in the treatment of metastatic lymph nodes in Stage IV squamous cell carcinoma of the head and neck.

Fast neutron radiation therapy. Results of phase III randomized trials in head and neck, lung, and prostate cancers

The results of phase III trials comparing neutrons to photons for head and neck squamous cell cancers, non-small cell lung cancers, and prostate adenocarcinomas are reviewed, with emphasis given to the most recent U.S. National Cancer Institute sponsored randomized clinical studies in which fast neutrons were delivered using modern, hospital-based, high-energy, isocentric-capable cyclotrons. In locally advanced squamous cell head and neck cancers, neutrons showed no convincing advantage over photons. Fast neutron radiotherapy may have provided a therapeutic benefit in selected patients with inoperable non-small cell lung cancers. For locally advanced prostate adenocarcinomas, neutron therapy resulted in significantly superior clinical and histological loco-regional tumor control, which may translate to improved survival with additional follow-up. In general, severe late complications were more frequent with neutrons, especially in patients treated on older physics laboratory-based equipment. Even with modern state-of-the-art neutron generators, careful beam collimation and treatment planning are required to minimize side effects.

An evaluation of the results of neutron therapy trials

Twenty-five randomised clinical trials have been completed which have investigated the value of fast neutron therapy. The results of these trials are reviewed in terms of the reported rates of local tumour control and late morbidity. The trials have included patients with cancers of the head and neck region, brain, lung, pancreas, cervix, bladder and rectum. None has demonstrated neutrons to be advantageous compared with photons. Two trials of locally advanced prostate cancer have given conflicting results. A trial of mixed schedule therapy has demonstrated improved local tumour control and survival. A trial of neutrons alone for similar stage disease has not shown any therapeutic advantages. It is still claimed that salivary gland tumours may be more effectively treated by neutrons but the clinical trial results are not definitive. Late morbidity after neutron therapy is a persistent cause for concern and often has been unacceptably high. There is no convincing evidence that fast neutrons are either as safe or as effective in cancer control as photon therapy.

The relative cellular radiosensitivity of 30 human in vitro cell lines of different histological type to high LET 62.5 MeV (p-->Be+) fast neutrons and 4 MeV photons

It has been suggested that fast neutron therapy may have a role in the treatment of those tumours which lie within the most photon-resistant histological categories. A clinical radiobiological study by Battermann et al., however, did not support this hypothesis (Battermann, J.J. et al., Eur. J. Cancer 17: 539-548, 1981). Similarly, in a comparison of the intrinsic cellular radiosensitivity of 20 human in vitro cell lines with 4 MeV photons and 62.5 MeV (p-->Be+) neutrons, there was no correlation between RBE and photon sensitivity. However, because the range of histological cell types in this in vitro study did not include sufficient representatives of the most sensitive and resistant histological categories, it was not possible to examine the relationship between histology and the relative efficacy of fast neutrons compared with photons. The intrinsic radiosensitivity of a further 10 human in vitro cell lines has thus been measured and the results of all 30 cell lines used in a comparison of the relationship between relative neutron sensitivity and histology. These results together with those obtained by reanalysis of published data from a clinical study of the RBE of pulmonary metastases by Battermann et al. suggest that in the clinical situation, photon-resistant histology per se may not be a sufficient criterion for the choice of high LET irradiation and emphasize the need for predictive assays for individual tumours.

Neon ion radiotherapy: results of the phase I/II clinical trial

Neon ion radiotherapy possesses biologic and physical advantages over megavoltage X rays. Biologically, the neon beam reduces the oxygen enhancement ratio and increases relative biological effectiveness. Cells irradiated by neon ions show less variation in cell-cycle related radiosensitivity and decreased repair of radiation injury. The physical behavior of heavy charged particles allows precise delivery of high radiation doses to tumors while minimizing irradiation of normal tissues. In 1979 a Phase I-II clinical trial was started at Lawrence Berkeley Laboratory using neon ions to irradiate patients for whom conventional treatment modalities were ineffective. By the end of 1988 a total of 239 patients had received a minimum neon physical dose of 1000 cGy (median follow-up for survivors 32 months). Compared with historical results, the 5-year actuarial disease-specific survival (DSS5) and local control (LC5) rates suggest that neon treatment improves outcome for several types of tumors: a) advanced or recurrent macroscopic salivary gland carcinomas (DSS5 59%; LC5 61%); b) paranasal sinus tumors (DSS5 69%; LC5 69% for macroscopic disease); c) advanced soft tissue sarcomas (DSS5 56%, LC5 56% for macroscopic disease); d) macroscopic sarcomas of bone (DSS5 45%; LC5 59%); e) locally advanced prostate carcinomas (DSS5 90%; LC5 75%); and f) biliary tract carcinomas (DSS5 28%; LC5 44%). Treatment of malignant gliomas, pancreatic, gastric, esophageal, lung, and advanced or recurrent head and neck cancer has been less successful; results for these tumors appear no better than those achieved with conventional x-ray therapy. These findings suggest that Phase III trials using the neon beam should be implemented for selected malignancies.

High energy (42-66 MeV reactions) fast neutron dose optimization studies in the head and neck, thorax, upper abdomen, pelvis and extremities

Five hundred and fifty patients were entered into a set of dose-searching studies designed to determine normal tissue tolerance to high energy (42-66 MeV reactions) fast neutrons delivered in 12 equal fractions over 4 weeks. Participating institutions included: The Fermilab (66 MeV p+----Be), The University of Washington (50 MeVp+----Be), U.C.L.A. (45 MeVH-----Be), M.D. Anderson Hospital (42 MeVH-----Be), and The Cleveland Clinic (42 MeVp+----Be). Patients were stratified by treatment facility and then randomized to receive 16, 18 or 20 Gy for tumors located in the upper abdomen or pelvis, and 18, 20 or 22 Gy for tumors located in the head and neck, thorax or extremities. Following completion of the randomized protocols, additional patients were studied at the 20.4 Gy level in the head and neck, thorax and pelvis. Normal tissue effect scoring was accomplished using the RTOG-EORTC acute and late normal tissue effect scales. Acute Grade 3 + toxicity rates in the head and neck were 19% for 20/20.4 Gy and 20% for 22 Gy. Time adjusted late toxicity rates in the head and neck at 12 months were 15% for 20/20.4 Gy and 0% for 22 Gy. The 18 Gy treatment arm of the head and neck protocol was dropped early in the study after only two patients were accrued. For cases treated in the thorax, acute Grade 3 + toxicity rates were 6% for 18 Gy, 15% for 20/20.4 Gy and 7% for 22 Gy. Late toxicity rates at 12 months were 0% for 18 Gy, 11% for 20/20.4 Gy and 18% for 22 Gy. Acute Grade 3+ toxicity rates in the upper abdomen were 0% for 16 Gy, 8% for 18 Gy and 12% for 20 Gy. There were no Grade 3 + late toxicities in the upper abdomen. In the pelvis, acute Grade 3 + toxicity rates were 0% for 16 Gy, 3% for 18 Gy and 3% for 20/20.4 Gy. Late Grade 3 + toxicities at 24 months were 20% for 16 Gy, 5% for 18 Gy and 24% for 20/20.4 Gy. In extremities, acute Grade 3 + toxicity rates were 7% for 20 Gy and 21% for 22 Gy while at 12 months, late Grade 3 + toxicity rates were 14 and 35%, respectively. The 18 Gy treatment arm of the extremities protocol was dropped early in the study after only two patients were accrued. Factors associated with normal tissue effects in addition to treatment dose are discussed.

Review of the clinical results of fast neutron therapy

The clinical results reported from the different neutron therapy centres are reviewed. Fast neutrons were shown to be useful in the treatment of locally extended inoperable salivary gland tumours (average local control rates 67% for neutrons compared to 28% for photons). For paranasal sinuses and some tumours of the head and neck area, especially extended tumours with large fixed lymph nodes, neutrons could also be of interest. By contrast, the results obtained for brain tumours were, in general, disappointing. Neutrons were shown to be beneficial in the treatment of well differentiated soft tissue sarcomas, as well as to bone- and chondrosarcomas. For locally extended prostatic adenocarcinoma, a RTOG randomized trial gave local control rates of 77% for mixed schedule compared to 31% for photons. Neutrons could be useful also for palliative treatment of melanomas. Further studies are needed in order to evaluate the benefit of fast neutrons for other localizations such as uterine cervix, bladder and rectum. It can be concluded that fast neutrons could be of interest for about 10% of the radiotherapy patients, but it is likely that the new high-energy hospital-based cyclotrons will further extend the indications of neutron therapy. However, patient selection remains one of the main problems and there is a need for development of individual predictive tests.

Fast neutron therapy for soft tissue sarcoma

Seventy-nine patients with soft tissue sarcoma were treated with fast neutron therapy at the Hammersmith Hospital, MRC Cyclotron Unit. Sixty-six of these, treated between 1971 and 1983 were assessable. The histology was reviewed and graded in 82% of cases and tumors divided into groups according to maximum diameter. In sixteen patients, who were irradiated following complete macroscopic removal of tumor, there was 94% local control and 86% survived 5 years. Of the 50 patients who had gross tumors present 62% were greater than 10 cm in diameter, and 20 were recurrent after previous radiotherapy or surgery or both. Sixty-eight per cent of gross tumors completely regressed and local control was 52%. The main cause of death was metastatic spread, and median survival was 63 months for Grade 1 patients, 9 months for Grade 2, and 7 months for Grade 3. Thus, there was a significant advantage to patients with Grade 1 tumor but little difference between Grades 2 and 3. Twenty-seven patients experienced late complications of treatment, 67% of which involved the skin predominantly and were related to the low energy of neutrons used. Seventeen of the 27 had received previous radiotherapy. Neutron therapy given in this dose and fractionation produced a higher local control rate than photon therapy, but complications were more frequent. Since these mainly involved the skin a lower level of complications may be anticipated using higher energy neutrons which will have a more even distribution of dose and lower skin dosage. Forty-eight per cent of patients developed metastatic disease, indicating the need for effective systemic therapy, especially in Grades 2 and 3 tumors.

Fast neutron therapy for squamous cell carcinoma in the head and neck region: results of a randomized trial

A randomized trial of fast neuron therapy compared with 4MV photons for patients with head and neck cancer is reported. One hundred and sixty-eight patients were recruited between 1977 and 1984. The minimum follow-up is 2 years. Three patients were withdrawn before treatment began. Eighty-five were allocated to neutron therapy and 80 to receive photon therapy. All patients had squamous cell cancers in one of four primary sites: oral cavity, oropharynx, larynx, and hypopharynx. Local tumor control was similar in both groups: 44.7% after neutrons and 45.0% after photons. Salvage surgery was performed on 18 patients in each treatment group for residual or recurrent cancer. Acute radiation reactions of the mucous membranes were significantly more severe after photons. The number of patients with serious late reactions was greater after neutron therapy but the difference was not statistically significant. There were six deaths related to late morbidity after neutron therapy but none after photon therapy. Survival was better after photon therapy but the difference compared with the neutron group failed to reach statistical significance. When intercurrent deaths are excluded, the difference is less marked. Photon therapy was clearly better in terms of disease-free survival giving a 2-year local disease-free rate of 41.3% (s.e. 5.5%) compared with 29.4% (s.e. 4.9%) after neutrons.