Clinical radiobiology normal tissue and fractionation Categories of

Clinical radiobiology normal tissue and fractionation

Categories of factors determining the severity of normal tissue damage • Controllable factors – Refers to total dose and irradiated volume – More severe and more likely to occur as the dose is increased – More frequent as the size of the field increase – Fractionation schedule is important • Uncontrollable factors – Age, clinical status, concurrent disease, genes, lifestyle

Early reactions ~ weeks after treatment start

Late reactions ~ months to years after treatment start teleangiectasi

Dose response curve The radiosensitivity of individuals is Gaussian distributed Dose response curve

Normal tissue tolerance Theoretical concept – difficult to put into clinical practice

Normal tissue tolerance

Steepness of dose response curves Is often given for the 37% and/or 50% level A 1% dose increment ( D) from a reference dose yields an increase in response equal to percentage points

Steepness of dose response curves

Gaussian distribution of radiation sensitivity Gaussian distribution of technical and dosimetrical factors Uniquely described by two parameters: - gradient at 50% response level D 50% - dose to achieve 50% response Dose response curve biolo > clin Clinical dose response curve Pettersen M et al Radiotherapy and Oncology 2008

Quantification of normal tissue damage A. Visual scoring

Quantification of normal tissue damage B. Assays of tissue function • Example: – Lung function – Urination frequency – Stool frequency – etc.

Fractionation

Total dose Strandqvist plot, irradiation of skin Number of fractions Ellis suggested the following equation (1969): D = NSD N 0. 24 T 0. 11 NSD – Nominal standard dose where D is total dose, N is the number of fractions, T is total treatment time

Isoeffect curves

Isoeffect curves • Different tolerance level for various tissue • Late-responding tissue have a steeper curve • Early-responding tissue is less sensitive to changes in fraction size

Mathematical modeling – the LQ-model • Assumption: – The cell is inactivated when both DNA strands are damaged • DSB can be produced a single track or by two close SSBs produced by independent tracks • Both these events are random and rare and can therefore be described by Poisson statistics • The probability of there being no such events, that is, the surviving fraction, S, of cells, is given by: Where p is the mean number of hits per cell

Mathematical modeling – the LQ-model • For single particle events, p is linear function of the dose, D • This means that the mean number of hits per cell can be expressed as α·D, and thus • where α is the average probability per unit dose that such single-particle event will occur

Mathematical modeling – the LQ-model • The mean probability of one particle causing damage in one arm of DNA in any specific cell is linearly proportional to the dose, D, and this will also be the case for the second particle • This means that the probability of both events occurring is D 2, so the probability of no such event will be • where β is the mean probability per unit square of the dose

Mathematical modeling – the LQ-model The overall LQ equation for cell survival is therefore: - letal damage - sub-letal damage The cell survival curve is blue and the initial slope is shown in red

Mathematical modeling – the LQ-model Of special interest is the dose at which the log-surviving fraction for -damage equals that for -damage:

/ -ratio represents the curviness of the curve The higher the / -ratio – the straighter the curve, such cells exhibits considerable irreparable damage and/or little repair (high and/or low ) Typical for cancer cells and early reacting tissue (high proliferation)

/ -ratio represents the curviness of the curve lower / -ratio – larger shoulder and more bending, such cells exhibits little irreparable damage or high capacity of repair (low and/or high ) Typical for late reacting tissue (low proliferation)

Fractionation • By full repair between fractions (at least 6 hours) the should will be repeated • The therapeutic gain will be improved • Late reacting tissue is more sensitive to fractionation

Mathematical modeling – the LQ-model The survival after fractionated treatment will be given by: Where n is the number of fractions

Sensitivity for fractionation Typical / -values for tumour tissue are approx. 10 Gy

Sensitivity for fractionation Typical / -values for late reacting tissue around 3 Gy

Sensitivity for fractionation Typical / -values for early reacting tissue around 3 Gy, i. e. the same as tumour tissue

LQ-model Effect = E = -log S 2 -( • d+ • d ) • n S=e E = ( ·d+ ·d 2)·n = ( + ·d)·D n – number of fractions with dose d, D – total dose TE = E/ = ( / +d)D BED = E/ = [1+d/( / )]·D

2 Gy equivalent dose, EQD 2 There is a lot of clinical experience where dose per fraction has been 2 Gy. Therefore, a non 2 Gy scheme is often expressed in 2 Gy equivalent dose, EQD 2. TE = (α/β + d)·D = (α/β + 2)·D 2 Gy

Sensitivity for fractionation Trough fractionation therapeutic gain can be improved Effect per unit dose / =2 -3 M. Saunders et al, Lancet Oncology, 2001 / =10

Models to describe survival curves

Combination of A and D The Universal Survival Curve (USC) Park et al 2008

Fractionation schedules • Conventional fractionation: one treatment per day á 2. 0 Gy, 5 days per week • Hyper fractionation: dose per fraction less than 2. 0 Gy, for the rest as above • Hypo fractionation: dose per fraction larger than 2. 0 Gy, for the rest as above • Accelerated fractionation: Several fractions per day; shorter total treatment time

Fractionation schedules • Some combinations: – Hyper fractionated, accelerated treatment: combination of hyper fractionation and shorter total treatment time – CHART: ’contineous hyper fractionated accelerated radiotherapy’; 3 fractions per day, each fraction had a lower dose than 2. 0 Gy. – Concomitant boost: hypo fractionated treatment of GTV, conventional fractionation of CTV. –. . .

Fractionation schedules Hyper fractionation gives a better therapeutic ratio, and thereby allows dose escalation. Tumour control Late reactions EORTC hyperfraksjoneringsstudie for pharynx-cancer. Same level of late reactions will occur, while the level of tumour control expects to increase compared to a conventional schedule.

Effect of overall treatment time

The volume effect