Search

We tested three radiobiologic models, the nominal standard dose (NSD), the biologic index of reaction (BIR), and the linear quadratic (LQ) models to determine which best predicts normal tissue injury in radiation therapy. Clinical data for radiation myelopathy, rib fracture and pericardial effusion were used for all three models to predict injury. We assumed that on the average injuries occurred at higher equivalent doses of radiation than were received by patients who were not injured. We used a t-test to determine whether there were in fact significant differences in the mean values of the equivalent doses among the injured and non-injured. The means were calculated for the four sets of injury by the three models. For the LQ model it was necessary to choose a value for the parameter α β ; the results were not sensitive to the choice over the range of 1 2 to 12 Gy. None of these models showed a significant difference between injured and non-injured patients for all four sets of data. The BIR model showed significant differences in three sets, the LQ model was significant in two and marginally significant in one set, and the NSD was significant in two sets. This analysis illustrates therefore, that the linear quadratic model can be adopted for analysis of clinical data with results that are no worse and possibly better than the NSD model.

The nominal standard dose concept is used to define the total biological dose or cumulative radiation effect at a point for treatment regimens employing two or more radiation modalities. Various clinical applications are illustrated. A convenient nomographic method for determining the total biological dose is presented.

The nominal standard dose depends mainly upon the skin reaction, the field size, and the location. In particular, different degrees of skin reaction cause variations in the “tolerance dose” up to a factor of 2. The influence of these parameters must be taken into account carefully in order to attain valid conclusions. The NSD concept has been verified by re-evaluation of Strandqvist's investigation. An exponent of 0.28±0.04 has been calculated, which is at the lower limit of Ellis' value (0.35). The use of the NSD formula may fail for radiation regimens which are not bound to skin reaction. Several radiation schemes in the treatment of lung cancer exhibit worse tumour control though NSD values are higher. For this a strong correlation between survival rate and fractionation frequency has been found.

Sixty-eight patients with varying degrees of radiation induced bone damage were studied in detail with respect to the absorbed doses in soft tissue of bone and the time-dose-fractionation treatment schemes employed.The cases were divided into 2 large categories of bone damage. The first category included 52 cases of predominantly irreparable injuries; the second included 16 cases of reparable damage.Average values of the NSD (Nominal Standard Dose) required for irreparable osseous injury ranged from 2,120 rets to 3,200 rets (roentgen equivalent therapeutic) depending on site. The average NSD value for reparable osteopathy was in the range of 1,310 nets.Analysis of iso-effect dose curves indicated varying degrees of dependency on the number of fractions on over-all treatment time for the production of radiation osteitis.Preventive measures such as careful treatment planning, utilization of supervoltage irradiation, and avoidance of trauma and infection are stressed.

Human tumors grown subcutaneously in the thymus deficient nude mice are used as a model system to compare the effectiveness of clinically relevant irradiation regimes. Procedures for local irradiation of the tumors with 60Co radiation are described in detail. A preliminary test of the regression of two malignant melanomas, following three clinically relevant fractionation schedules, is performed. The overall treatment time and the nominal standard dose (NSD) are kept equal for the irradiation regimes chosen.

Precision dosimetry and computer augmented optimization of isodose patterns have become standards of good practice in external beam radiotherapy, but conventional dose expressions and multifield isodose patterns present inadequate information about dose-fractions. From the Walter Reed disaster following megavoltage treatment of 256 testicular malignancies in the 1940s to the present time, the practice of treating alternating fields accounts for most of the dorsal myelitis, heart damage and gastrointestinal injuries outside the pelvis. Our most recent publications and scientific meetings indicate little progress understanding bow the treatment of three or four converging fields daily, 5 days a week, produces radiation effects different from those obtained when the same tumor dose is given in the same time with opposed or converging fields treated in succession. The formulation of relative radiation effect (RE) ratios is described, along with manual methods, and computer (PC-12) techniques for producing iso effect patterns. A variety of multifield isodose patterns and fractionation schedules have been converted into iso effect diagrams to provide comparison and discussion of RE distributions. No specific nominal standard dose (NSD) tolerance dose figures are needed in the calculations. Visual comparison of RE distribution makes it easy to select a plan least harmful to a critical organ.

Various dose-time treatment plans have been used to obtain long duration pain relief in patients with metastatic bone disease. Very little has appeared in the literature evaluating the relationship of dose and fractionation to initial, delayed and permanent bone pain relief. At the Swedish Hospital Tumor Institute, 152 treatment fields in 110 patients were evaluated, with a clinical follow-up in many of over five years. Those treated at lower total doses with less fractionation achieved the same quality and duration of pain relief as higher doses. Treatment plans also were compared using the Ellis method of nominal standard dose.

The technique of treating only one prescribed field daily in a multi-field setup produces an increased biological effect upon the normal tissues. In a consecutive group of 18 patients with Hodgkin's disease all treated employing extended opposed fields on a 60Co unit and treating only one prescribed field daily, 8 patients have delayed complications attributable to radiotherapy. All 18 have been followed for more than four years. The concept of Nominal Standard Dose (NSD) has been utilized in this study to compare the dosages in critical areas, i.e., the heart and spinal cord, relative to the tumor dose. Tolerance was exceeded to a greater extent in the patients with complications.

The significance of treating all fields or only one prescribed field at each radiotherapy session is evaluated using the concept of nominal standard dose as well as data from a mammalian cell survival curve. Analysis with either system demonstrates greater homogeneity of biological effect within the treatment volume when all prescribed fields are administered at each treatment session. The importance of this effect varies with the clinical situation.

An analysis of selected papers in which the Nominal Standard Dose (NSD) concept is used reveals that about 50% of them contain errors. These mistakes are both conceptual and mathematical in nature and average plus or minus 7%. Typical errors and methods to eliminate them are discussed. A table is provided for interconversion of NSDs, partial tolerances, and time, dose and fractionation (TDF) factors.

A study was made of 17 patients with chemodectoma of glomus jugulare and 23 patients with carcinoma of the auditory canal, middle ear or mastoid from the standpoint of the relationship between time dose fraction values and the occurrence of temporal bone osteoradionecrosis. Osteoradionecrosis occurred in 1 of 17 patients with chemodectoma (6%) and 6 of 23 patients with carcinoma (26%) following curative radiation therapy. Of 12 patients who received radiation Nominal Standard Dose (NSD) value more than 2000 ret or time-dose-fraction value (TDF) 120 to the major portion of the temporal bone, seven developed osteoradionecrosis. No patient receiving a radiation volume less than this magnitude had such complications. The optimum doses for the chemodectoma of the glomus jugulare and the carcinoma of the auditory canal, middle ear or mastoid are suggested.

The nominal standard dose concept proposed by Ellis 1967) for comparison of various treatment regimes has been widely accepted. The nominal standard dose is associated with treatment schedules which lead to full tolerance of tissue. Using this idea, Kirk et al. (1971) have proposed the use of Cumulative Radiation Effects (CRE) at the partial tolerance level. Orton and Ellis (1973) have proposed time, dose, fractionation factors for dealing with treatment schedules leading up to and including tolerance. A question, therefore, arises which of these two concepts should be used for dealing with problems of radiotherapy at the present juncture.

The relationship between total dose, number of fractions (N) and over-all treatment time (T days) is of great importance in radiotherapy when any non-standard fractionation schedules are considered. The use of the formula by Ellis (1969) relating these qualities for X and γ rays, i.e. where N.S.D. = nominal standard dose), is increasing, partly because of its simplicity and partly because of the availability of a slide rule adjusted to fit its parameters (Winston, Ellis and Hall, 1969).

The local control and complications after radical irradiation were analyzed in 203 patients with supraglottic cancer, using time-dose scattergrams and nominal standard dose (NSD). A dose-response curve was obtained for early (T-1, T-2) tumors but was not demonstrated in advanced neoplasms (T-3, T-4). All nodes less than 2.5 cm in diameter were controlled with 7,000 rads. Of the early lesions 89% were controlled with 7,700 rads in six weeks. Over 85% of the involved nodes were controlled when the primary was cured. The incidence of major complications did not exceed 7%.

The nominal standard dose concept was used to evaluate the difference in biological effect at the depth of maximum buildup (dm) and at the midline for parallel opposed fields. The evaluation was for all field sizes for 60Co, 4, 6 and 22 MV x-rays and all separations up to 24 cm. The biological effect is almost always greater at dm than at the midline and for a given midline dose is much greater if only one field is treated daily as compared with both fields treated daily.

Tables with instructions and examples are presented which allow calculation of NSD values for complex treatment schedules with gaps or changes in fractionation pattern. These are based on a previously derived formula, and eliminate the need for a computer to calculate these values.

An analysis of the data of Hopewell and Berry concerning the tolerance of the pig kidney to radiation, shows that the Nominal Standard Dose (NSD) concept provides an adequate indicator of renal tolerance within the range of fractionation and overall time in common use in clinical radiotherapeutic practice.

A comparison was made of the response of the KHT sarcoma, transplanted into the thighs of syngeneic C3H mice, and the response of the normal tissues of the irradiated leg to two fractionation regimes; six fractions in two weeks and ten fractions in two weeks. The response of the tumour, which was measured by the delay in growth to a fixed size, was found to be consistent with a fractional difference between the two regimes of approximately 11 per cent for dose fractions of between 200 and 600 rads. For the normal tissues, early skin reactions and deformity 48 weeks after irradiation were measured, and the mean fractional difference between the two regimes was 12 per cent for the early and 13 per cent for the late reaction. The values of the “fraction number” exponent in the nominal standard dose equation of Ellis were calculated for both the tumour and normal tissues and found to be consistent with the value of 0·24.

An analysis of early published multifraction orthovoltage human acute skin irradiation tolerance isoeffect doses is presented. It indicates that human acute skin radiation reactions may result from the repetition, with each dose fraction, of a cell survival curve of the form: S = e- −(αD + βD 2 ) The analysis also shows no need for an independent proliferation related time factor for skin, for daily treatments of six weeks or less in duration. The value obtained for the constant β/α for orthovoltage irradiation from these data is 2.9 × 10 −3 rad- −1 ' for the cell line determining acute skin tolerance. A radiation isoeffect relationship, based on the quadractic cell survival curve, is introduced for human skin. This relationship has some advantages over the nominal standard dose (NSD). First, its use is not restricted to tolerance level reactions. Second, a modification of the relationship, which is also introduced, may be employed in the selection of doses per treatment when irradiation dose fractions are administered at short intervals where repair of sublethal injury is incomplete.

Fields on the flanks of domestic swine were irradiated with 250 kVp X rays in courses of one, six, or 30 fractions. The acute skin response was scored quantitatively using a numerical scheme. Late skin tolerance, up to one year after exposure, was determined by the presence or absence of skin necrosis in the irradiated fields. The Nominal Standard Dose was a poor predictor for early skin response; an approximately 30 per cent higher equivalent dose was required when administered in six fractions over 18 days to produce either dry or moist desquamation, than was required when given in 30 fractions over 39 days. The difference was even greater when compared with single radiation doses. Late skin tolerance was not closely correlated with the level of early response, but skin necrosis was well predicted by the NSD formula. These data in the pig correlate well with clinically determined late skin tolerance in three large radiotherapy centres. For a 64 cm2 skin field, however, skin tolerance in the pig...

Based on published clinical results of radiotherapy, a formula is suggested which relates total dose, number of fractions and overall treatment time to a quantity termed ‘Nominal Standard Dose’. This quantity represents the biological effect of a given treatment regime. Using this concept it is possible to compare various treatment schedules that involve different fractionation patterns and various overall treatment times. The evidence upon which the idea is based, and also its use in routine clinical practice, are discussed.

The use of the NSD concept in practical radiotherapy is simplified by the introduction of time, dose, and fractionation (TDF) factors, which are proportional to partial tolerances, but are not dependent upon any specific NSD value. Detailed examples are given of the utilization of TDF factor tables, which are simple to use, general in application, and which help to ensure the correct interpretation of the basic NSD concepts. Computers, slide-rules, or even log tables, are not required. The concept of nominal standard dose (NSD), introduced by Ellis (1967), has achieved widespread acceptance because, in spite of its obvious limitations, it enables radiotherapists to compare and contrast treatment regimes which differ in time, dose, and fractionation patterns. In a recent survey (Orton, 1972), a number of experts, chosen because they had been authors of papers on this subject, were asked to solve a problem involving a comparison of different radiotherapy regimes. The Ellis NSD concept was misused by more than half of the experts who attempted to solve the problem by this method. This resulted in dose errors ranging from +5 to +1 9 per cent with an average of +1 1 per cent. Evidently the practical application of the idea is not entirely understood and some simplification is desirable. The purpose of this paper is to introduce a set of tables which are simple to use and general in application, and which help to ensure correct interpretation of the basic NSD concepts. NOMINAL STANDARD DOSE From a consideration of iso-effect relationships in clinical radiotherapy, Ellis (1967) proposed that the tolerance dose for normal tissues (D rads) could be related to the overall treatment time {T days) and the number of fractions (N), by the expression

1. The criticisms adduced by Liversage of the N.S.D. concept and the Ellis formula are based on misunderstanding and insufficient consideration of the basic reasoning and evidence. 2. In clinical practice, by providing a single figure representing biological effectiveness, the concept and the formula are useful for teaching and comparison of treatment schedules besides being safe for prescribing. 3. Virtual coincidence of isoeffect values for normal tissue (N.S.D.) (Table I) and for Hodgkin's disease (tumour S.D.) (Table V) support the concept and the formula. For most other tumours the tumour effect is incidental to normal tissue tolerance. 4. Optimum schedules for various types of tumour might be possible to derive when more is known of the mitotic cycle times of the affected normal and malignant cells underradiation treatment. 5. Continual testing and sifting of data are desirable to verify the concept and to adjust the exponents of the formula.

The methods of use of the Ellis Nominal Standard Dose equation are described; this equation relates dose, number of fractions and time of treatment to a parameter representing normal tissue tolerance. The equation is in an index form and is inconvenient to solve. A specially constructed slide-rule calculator is described which facilitates the evaluation of the treatment parameters.

It is only because I have such a high regard for Dr. Ellis's concept of establishing a unit of biologically effective dose, that I have taken the trouble to criticize his suggestions in such detail. The better the idea the more carefully it should be considered, and this perhaps is why some of the criticism in my article, “A critical look at the ret” (Liversage, 1971) appears to Dr. Ellis to be of a trivial nature. Dr. Ellis replies to my criticism in his paper “Nominal standard dose and the ret” (Ellis, 1971) and on page 102 of this paper he disagrees with my findings on five counts which I wish to discuss further in the light of his comments and in the same numerical order.

The St Thomas' experience in post-mastectomy radiotherapy (Bates, 1975; Bates & Peters, 1975) became one of the cornerstones in demonstrating that the nominal standard dose (NSD) formalism (Ellis, 1969), in general, underestimates the deleterious effect of large dose fractions on late-responding normal tissues. It is therefore with interest that we read the 10- year update of the treatment results in this series in the July 1988 issue of BJR (Bates, 1988). There are, however, several points where this paper is at variance with the experience from other clinical radiobiological analyses published in the last few years. We would like to discuss briefly some of these.

When prescribing radiotherapy dosage regimes in unusual circumstances, a therapist must usually resort to the use of such concepts as the nominal standard dose (Ellis, 1969) or other models, whereby the effective dose, distributed over a period of time in discrete fractions, may be calculated. Such unusual circumstances may include the establishment of new treatment techniques, the need for reduced numbers of fractions due to other treatment factors such as hyperbaric oxygen, anaesthesia, etc., or when holidays or machine breakdowns cause patterns to be disrupted.

The concepts of nominal standard dose (NSD) and cumulative radiation effect (CRE) are now well accepted in radiation therapy. The unit of NSD is ret (rad equivalent therapy) and the number of rets for various fractionation schedules have been published (Phelps and Phelps, 1974). With the introduction of SI units in radiology and the requirement that the special units of radiation be phased out within a period of ten years, it will be necessary that the unit ret should also be replaced.