A comparison of dose-volume constraints derived using peak and longitudinal definitions of late rectal toxicity

Accurate reporting of complications following radiotherapy is an important part of the feedback loop to improve radiotherapy techniques. The definition of toxicity is normally regarded as the maximum or peak (P) grade of toxicity reported over the follow-up period. An alternative definition (integrated longitudinal toxicity (ILT)) is proposed which takes into account both the severity and the duration of the complication.

In this work, both definitions of toxicity were used to derive dose-volume constraints for six specific endpoints of late rectal toxicity from a cohort of patients who received prostate radiotherapy in the MRC RT01 trial. The dose-volume constraints were derived using ROC analysis for 30, 40, 50, 60, 65 and 70 Gy.

Statistically significant dose-volume constraints were not derived for all dose levels tested for each endpoint and toxicity definition. However, where both definitions produced constraints, there was generally good agreement. Variation in the derived dose-volume constraints was observed to be larger between endpoints than between the two definitions of toxicity. For one endpoint (stool frequency (LENT/SOM)) statistically significant dose-volume constraints were only derived using ILT.

The longitudinal definition of toxicity (ILT) produced results consistent with those derived using peak toxicity and in some cases provided additional information which was not seen by analysing peak toxicity alone.