16. Cervix cancer - The GEC-ESTRO Handbook of Brachytherapy

Cervix cancer

20

THE GEC ESTRO HANDBOOK OF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 01/09/2023

when using radiobiological models to support deviation from commonly used clinical schedules as there is limited knowledge about actual repair half-times. For HDR treatments, the fraction size can be modified as the therapeutic window between tumour and OAR EQD2 is wider with smaller fraction doses. In many centres, HDR BT is delivered in 4 fractions, often delivered in two separate applications. The number of HDR fractions may be reduced to 3 fractions without compromising target coverage or increasing dose to normal tissues [51], particularly in low-risk patients with small or good responder tumours. It may also be desirable to deliver more than one fraction of HDR BT using the same treatment plan in the same application to reduce the resources required [52]. At all times, it is important that changes in fraction size and timing of BT using radiobiological calculations should respect the limitations of the radiobiological models involved and are based on published clinical experience. 10.4 Intra- and inter-fraction variations In IGABT, there is typically a 1-5 h delay between imaging and dose delivery to allow for contouring and dose planning. Applicator and organ movements (e.g. filling status of bladder and rectum) during this period will result in deviations between the planned and delivered dose. Moreover, when delivering several HDR fractions per insertion, MR or CT imaging may not be performed for each individual fraction resulting in inter-fraction variations. With PDR BT, there is the additional risk of movement during and between pulses. Intra- and inter-fraction organ movement has less impact on tumour dose than on OAR dose as the applicator is fixed within the tumour. Some centres utilise MRI or CT for the first BT fraction only due to resource limitations and succeeding fractions are treated with the dosimetry plan from the first fraction. While this is analogous to the practice for EBRT, EBRT treatment plans are generated to take into account inter-fraction uncertainties by application of a PTV margin. In BT, a PTV margin cannot be used due to the steep dose gradients. Moreover, it is unnecessary in a single insertion as the target tends to move together with the applicator. This is not the case, however, if only one imaging session is used for several BT insertions. Target shrinkage between subsequent BT fractions will result in higher target doses. The OAR doses may also be higher depending on their location. The applicator geometry and position may differ between insertions, particularly for combined IC/IS treatments; in these cases, it is recommended that images be obtained for each applicator insertion because the applicator geometry is not reproducible between insertions of needle applicators, and use of the same treatment plan is not desirable. In vivo dosimetry has been shown to be useful for source tracking to detect errors in connecting the applicators to the afterloader and to ensure correct delivery; this is especially important for prolonged PDR treatments [53]. 10.5 Recommendations for reporting The modern practice of cervical cancer BT requires significant resources for patient work-up, imaging, preparation, treatment planning, and delivery. However, the availability of the full range of resources varies dramatically throughout the world. ICRU Report 89 has therefore recommended different levels of recording and

reporting dependent on resource availability. • Level 1 describes the minimum requirements that should be followed at all centres for all patients. Different recommendations have been defined depending on the type of imaging used for planning BT i.e. radiographic or 3D volumetric (MRI, CT, US). • Level 2 indicates advanced standards for planning and treatment that allows more comprehensive and standardised exchange of information between centres. • Level 3 describes new forms of planning and treatment related to research and development for which reporting criteria are yet to be established. For each parameter, it is necessary to distinguish between the planning aim, which is the desired dose to the target and OAR, and the prescribed dose, which is the achieved dose in an individual patient after optimisation. The prescribed doses in an individual patient can be correlated to local control and patient morbidity. For a patient cohort, the prescribed doses can be reported using descriptive statistics (mean, median, range, standard deviation) and also as the proportion meeting the planning aims as a quality indicator for the service. 11.1 During treatment Cervical cancer BT is a complex process and careful monitoring of patients is necessary for optimal treatment delivery. Patients are required to remain immobile for the duration of treatment to minimise the risk of applicator movement. Tight packing of the vagina after applicator insertion is crucial to prevent displacement. For HDR treatments, in particular, it is important to check for applicator displacement prior to each fraction e.g. by comparing against marks on the legs. Anti-embolic stockings and anti-coagulant prophylaxis are recommended if patients are immobilised for >24 hours. Constipating agents may also be necessary to minimise bowel action during immobilization. Pain control can be problematic requiring use of epidural or patient-controlled analgesia for the duration of treatment. Adequate analgesia is particularly important for applicator removal. After applicator removal, it is important to check for evidence of bleeding, in particular, when IS needles have been used. Any bleeding can usually be controlled by applying direct pressure on the source of bleeding in the cervical or paracervical area. equipment should always be on standby. While applicators are usually removed in entirety, it is nevertheless important to check for retained components. After removal of IC/IS applicators, it may be prudent to check the haemoglobin level to detect rare clinically occult intra-abdominal bleeding. 11.2 Follow-up All centres should monitor the outcome (tumour control and late morbidity) of their own patients to ensure that their results are commensurate with published data. Analysis from the Retro EMBRACE study [54] showed that 40-50% of all recurrences occurred in the first year with a further 20-30% occurring in the second year. Overall, 85.2% of recurrences occurred in the first 3 years and 94.5% within 5 years. Although local, regional, and PAN failure tended to plateau after year 3, systemic failure continued to occur for up to 10 years. A small number of late recurrences of all types were seen after 10 years. 11. MONITORING