D. Phil

Residual, unresectable or recurrent colorectal cancer: external beam irradiation and intraoperative electron beam boost ± resection

While combinations of external beam radiation (XRT) and surgery decrease pelvic recurrence and improve survival in the subgroups with residual disease (postop XRT) or initially unresectable disease (preop XRT), local recurrence is still unacceptably high, and survival could be improved. In view of this, pilot studies were instituted at Massachusetts General Hospital in which 32 patients received the standard previous treatment of external beam irradiation and surgery but in addition had an intraoperative electron beam boost of 1000-1500 rad to the remaining tumor or tumor bed. For the 16 patients who presented with unresectable primary lesions, the addition of intraoperative radiotherapy has resulted in a total absence of local recurrence with a minimum 20 month follow-up, and survival rates are statistically better than for the previous group treated with only external beam irradiation and surgical resection. In the group with residual disease, again there have not been any local recurrences in the 7 patients who received all treatment modalities versus 54% and 26% for the group with gross and microscopic residual treated with only external beam techniques. The remaining 9 patients presented with recurrent unresectable lesions--3 are alive (2 NED) at greater than or equal to 3 years.

Relative biological effectiveness of modulated proton beams in various murine tissues

The relative biological effectiveness (RBE) of proton beams produced by Harvard University 160 MeV synchrocyclotron was studied in various murine tissues. Reference radiation was Cobalt-60 gamma-rays from a teletherapy unit at the Massachusetts General Hospital. Animals were C3Hf/Sed mice derived from our defined flora mouse colony. Test tissues are: lens, lung, testes and tail vertebrae. The RBE of the third generation isotransplants of a spontaneous mouse mammary carcinoma was also investigated. The proton and Cobalt-60 irradiations were carried out simultaneously by 2 teams. The dose response curves obtained for testes weight loss and growth stunting of tail vertebrae indicated that the RBE for our protons was independent of radiation dose in the range of 0.4 to 16 Gy. This finding was identical to our previous studies of the murine fibrosarcoma, skin and small intestine. The RBE values for lens and lung tissues were obtained by determining radiation dose to result in a complete cataract in half the irradiated eyes in 210 days and a 50% mortality in 180 days respectively. We have studied proton RBE in 7 normal tissues and 2 tumors including previously reported results. The RBE values for these tissues were found to fall between 1.09 and 1.32. No significant differences in the proton RBE were found between the several normal and tumor tissues studied.

Fibromatoses: From postsurgical surveillance to combined surgery and radiation therapy

The results of management of two groups of patients with musculoaponeurotic (desmoid tumors) and plantar fibromatoses seen at Massachusetts General Hospital (MGH) during the period 1970-1985 are examined: (a) 26 patients who had had surgical resection for their primary fibromatosis but whose surgical margins were positive and who received no further treatment; and (b) 24 patients who were treated for their primary or recurrent fibromatosis by radiation alone or combined with surgery. For the 26 patients who were only observed, despite the positive surgical margins, 9 have recurred; the actuarial continuous local control rate at 5 years was 68% (a median follow-up of 70 months). Five patients had gross disease left after surgery and all of them failed. Seventeen of 21 patients who had grossly complete resection have local control; the four failures have been salvaged. This result supports the rationale for a no treatment but a thorough and close follow-up policy for patients with positive margins after grossly complete resection of a primary desmoid or fascial fibromatosis. There is no risk of metastasis in these patients and hence the effort toward a conservative policy which defers radiation merits interest and further study. Of the second group, 23 patients were treated for gross disease and one patient for microscopic disease after surgical resection. All of the 10 patients who were treated for primary desmoid tumor have local control. Among the 14 recurrent desmoid tumors there have been five local failures, after treatment by radiation alone or radiation + surgery. Three patients treated by radiation alone are currently scored as incompletely regressed tumors. Accordingly 16 of the 24 patients are scored as local controls without evidence of disease and 19 of the 24 are scored as local control (complete response or partial but stable response).

Twice-a-day radiation therapy for supraglottic carcinoma

This is a report of a group of 106 patients with supraglottic carcinomas treated by the twice-a-day radiation therapy program at the Massachusetts General Hospital from October 1979 through April 1984. The program consisted of 1.6 Gy per fraction, 2 fractions a day, 5 days a week for a total of 64 Gy with 2 weeks rest after the twelfth b.i.d. day. The local control of this group of patients was compared to that of 79 patients treated by the conventional once-a-day program with a daily fraction of 1.8 Gy for a total of 65 Gy, at the same institution during the 4 years immediately prior to the b.i.d. program. The 3 year actuarial local control rate for the entire group following the twice-a-day program was 76% as compared to 50% after the once-a-day program. The difference was significant, p = 0.001. For the T1 and T2 lesions, the corresponding rates were 88 and 63%, respectively, with a p value of 0.029. The rates for T3 and T4 lesions were 66 and 33%, respectively, p = 0.0037. The study indicated that the twice-a-day radiation therapy program as outlined is effective in treatment of supraglottic carcinoma, and is more markedly effective in advanced lesions. Late radiation effects are minimal and salvage surgery is possible for radiation therapy failures. To date, no patient developed radiation myelitis following the b.i.d. program with the dose to the spinal cord limited to 38.4 Gy in 2.5 weeks.

In vivo determinations of RBE in a high energy modulated proton beam using normal tissue reactions and fractionated dose schedules.

Abstract In preparation for a clinical trial of proton radiation therapy, we have performed a series of in vivo radiobiological experiments to determine the relative biological effectiveness (RBE) of protons as compared to 60 Co using normal tissue reaction in mice as the biological endpoint. The first assay studied was the survival of jejunal crypt cells in both the spread out Bragg Peak (SOBP) and the unmodulated portion (valley) of the proton beam. Experiments were performed using both single fraction and 20 fraction (3 hr between fractions) irradiations. These data demonstrate an RBE in the SOBP and the proton valley of 1.23 and 1.17 respectively for the 20 fraction experiment, and 1.19 and 1.13 for the single fraction experiment. These RBE values are only minimally dependent on the endpoint used for the analysis. The second assay was the acute skin reaction in the mouse thigh after 20 fractions of irradiation. Analyzing the average response from 11 to 30 days after irradiation the RBE was 1.13, and analyzing the dose for 50% of the animals to reach a score of 2.5 indicated an RBE of 1.14. Analysis of late effects by measurement of thigh epilation indicated an RBE of 1.14 while analysis of contraction was inadequate for an RBE estimate. These experiments indicate an RBE of approximately 1.20 for protons (compared to 60 Co) when analyzing mouse jejunum and approximately 1.13 when analyzing acute and late skin reaction. They also indicate that the RBE is relatively independent of fraction size in the range studied, and independent of whether the irradiation is performed in the center of a 10 cm wide SOBP or in the valley of the proton beam.

Evaluation of the clinical applicability of proton beams in definitive fractionated radiation therapy.

We report on the treatment of 317 patients treated either wholly or in part with proton beams at the Harvard Cyclotron Laboratory. These include: 130 patients treated for definitive radiation therapy of choroidal melanoma; 17 patients treated for tumors of the base of skull, cervical spine and cranium, which abut structures of the central nervous system (CNS); 23 patients treated for sarcomas of soft tissue and bone; 65 patients treated for carcinoma of the prostate; 14 patients treated for carcinoma of the rectum and anus; and 23 patients treated for squamous carcinoma of the oral cavity and oro-pharynx. Data on causes of failure and morbidity of treatment are presented. Overall the results are judged to be extremely encouraging. In particular, the treatment of the choroidal melanomas and sarcomas abutting CNS structures have clear clinical value, and the treatment of prostatic tumors and tumors of the head and neck are thought to be promising.

The response of two human tumor xenografts to fractionated irradiation. the derivation of α β ratios from growth delay, tumor control, and in vitro cell survival assays

A series of growth delay experiments was performed to derive alpha/beta ratios for two human neoplasms growing as xenografts in the hind limbs of NCr/Sed nude mice. The tumors were irradiated at 6 mm mean diameter under clamp-hypoxic conditions in one, two, four, or eight fractions, 2 fractions per day with a minimum intertreatment interval of 4 hr and a maximum overall treatment time of 3 days. The alpha/beta ratios derived for the high grade glioma U87 and the pharyngeal squamous carcinoma FaDu were 38 and 20 Gy, respectively. Comparably high values were derived from the same two tumors in a reanalysis of fractionated TCD50 data. The alpha/beta ratios were similarly high whether the TCD50 data were analyzed using the Full Effect plot or the Direct Method. For comparison, cell survival assays were performed on U87 and FaDu irradiated in vitro under plateau-phase, aerobic conditions. The alpha/beta ratios obtained were 9.2 and 15.0 Gy, respectively. Such high alpha/beta values suggest a therapeutic gain could result from the use of small doses per fraction.

Potential impact of improvements in radiation therapy on quality of life and survival

The NCI goal for the U.S. is to reduce the cancer mortality rate to one-half by the year 2000. Part of this improvement will be due to use of state of art treatment on a population wide basis and in addition, major improvements in efficacy of treatment. More effective radiation therapy will result in fewer local failures, increased survival, reduced treatment associated morbidity, and less frequent complications of treatment. The latter two aspects are very important to quality of life. If radiation combined with other modalities (e.g., sensitizers) were to become of greatly enhanced efficacy, then the number of patients requiring cystectomy, abdominal perineal resection, pneumonectomy, amputation, etc., would come down. The beneficial consequence would be fewer patients subjected to permanent ileostomy, colostomy, etc. The increase in survival if local failure were eliminated was estimated to be the decrease in local failure less the same loss in the new local controls due to DM as obtained in the local control patients after conventional treatment. For all sites, patients experience higher survival rates if they do not have local failure. For those sites where loss due to metastasis is not high, more effective local therapy would result in numerically impressive gains in survival.

Concurrent and subsequent tumors in the same host: A model to evaluate the host tumor interaction

Abstract There is evidence which demonstrates an immune rejection response (IRR) directed against tumor cells in some experimental and human tumors. In the case of FSa-1, a methylcholanthrene-induced fibrosarcoma of the C 3 Hf/Sed mouse, the IRR is manifested by a decrease in the dose of radiation expected to control half of the treated tumors (TCD 50 ) and an increase in the number of tumor cells expected to transplant the tumor in half of the transplanted recipients (TD 50 ) in immunized hosts. FSaI was transplanted simultaneously in the right and left legs of male and female C 3 Hf/Sed mice and each tumor was given 3750 rad when it measured 8 mm. in diameter (viz the TCD 50 value for 8 mm FSaI growing as one isotransplant per animal). The two tumors in any one animal usually responded similarly with either permanent regression or local recurrence after irradiation. In a second experiment, a group of animals received also concurrent tumors. This time an immunogenic FSaI was transplanted in one side and a non immunogenic mammary carcinoma (MDAH-MCaIV) on the opposite side. The tumors were irradiated with TCD 50 doses (3750 rad for FSaI and 6500 for the less radioresponsive MCa). The distribution of local control and recurrence was probabilistic; 50% of the animals exhibited one tumor destroyed and one recurring. In a third experimental group mice were subsequently transplanted with FSaI Mice who showed no recurrence after irradiation of their first FSaI showed a stronger tendency to reject the second implantation than those who previously had local recurrence and subsequent amputation. Retransplants were controlled more easily by irradiation in the group that had been successfully treated for the first transplant. These data indicate that within the inbred population of C 3 Hf/Sed mice there is a relatively broad distribution of capacities to react effectively against the FSaI challenge. This was not predicted by the delayed hypersensitivity skin reaction (DHSR) to the FSaI antigen.