Ultrahypofractionated Versus Normofractionated Sequential Boost After Whole-breast Radiation Therapy in Patients Treated With Breast-conserving Surgery for Breast Cancer (ULTIMO)

In the control arm, the boost of the radiation therapy will be delivered according to the standard of care fractionation and dosage scheme. This consists of 5 fractions of 2Gy each, adding up to a total dose of 10Gy.

All other treatments are not considered to be study-specific and are performed according to the standard of care treatment for breast cancer.

• normSEB

In the experimental arm, the boost of the radiation therapy is delivered in a ultrahypofractionated scheme. This consists of a single fraction of 6Gy.

All other treatments are not considered to be study-specific and are performed according to the standard of care treatment for breast cancer.

• ultSEB

Operationalisation:

BCCT.core is a software package for evaluating photographs of breasts after BCS. The 'Global cosmetic result' from the two frontal photos (arms up, and arms down) will be averaged and then used as the individual value of this measure. The results are scored on a 4-point Likert scale, coded as 0-3.

Analysis metric:

The change from baseline at 1 and 3YFU will be used. Negative values will be transformed to '1' to signal an inferiority event, positive or values of zero will be transformed to '0' to signal non-inferiority.

For comparisons and estimands, please refer to the SAP.

Method of aggregation:

The results will be reported as a contingency table reporting both frequencies and proportions of non-inferior and inferior results in each treatment arm.

Time point(s):

A baseline assessment is performed during the screening visit, followed by repeated measurements during the 1 and 3YFU (primary endpoint) visits.

Operationalisation:

The AIS is a tool for scoring breast cosmesis based on standardized photos. It comprises of 5 items, each comprising of a 5 point Likert-scale, coded as 1 to 5. The score is given based on the group of photos taken at a single time point. The scores of all items are summed to produce the TAS, which consequently ranges from 5 to 25. The AIS is used in an expert panel setting, where all scores, from each assessor are averaged per item.

A higher scores represents a more favourable cosmetic outcome.

Analysis metric:

AIS-TAS absolute values will be used for analysis.

Method of aggregation:

The mean, median, SD and IQR will be reported. For comparisons and estimands, please refer to the SAP.

Time point(s):

A baseline assessment is performed during the screening visit, followed by repeated measurements during the 1 and 3YFU (primary endpoint) visits.

Operationalisation:

The AIS is a tool for scoring breast cosmesis based on standardized photographs. It comprises of 5 items, each comprising of a 5 point Likert-scale, coded as 1 to 5. The score is given based on the group of photos taken at a single time point. The AIS is used in an expert panel setting, where all scores, from each assessor are averaged per item.

A higher scores represents a more favourable cosmetic outcome.

Analysis metric:

AIS-Symmetry score absolute values will be used for analysis.

Method of aggregation:

The mean, median, SD and IQR will be reported. For comparisons and estimands, please refer to the SAP.

Time point(s):

A baseline assessment is performed during the screening visit, followed by repeated measurements during the 1 and 3YFU (primary endpoint) visits.

Operationalisation:

The physician scoring scale (questionnaire, cfr. protocol) will be used to assess cosmetic outcome. The normalised average score of all items will be summed and used as outcome measure. This will be calculated as the score of an item divided by the amount of answering options minus 1 (k-1), averaged over all items. Some questions have the option 'not evaluable', if this answer is selected, then that item is not used in deriving the average score.

A higher scores represents a more favourable cosmetic outcome.

Analysis metric:

Absolute values will be used for analysis.

Method of aggregation:

The mean, median, SD and IQR will be reported. For comparisons and estimands, please refer to the SAP.

Time point(s):

A baseline assessment is performed during the screening visit. It is then assessed again at 1 year, and 3 years of follow-up after the last study treatment (LST).

Operationalisation:

The cosmetic evaluation questionnaire by Sneeuw at al. will be used to assess cosmetic outcome, as reported by the participant. The normalised average score of all items will be summed and used as outcome measure. This will be calculated as the score of an item divided by the amount of answering options minus 1 (k-1), averaged over all items.

A higher scores represents a more favourable cosmetic outcome.

Analysis metric:

Absolute values (calculated score) will be used for analysis.

Method of aggregation:

The mean, median, variance and IQR will be reported. For comparisons and estimands, please refer to the SAP.

Time point(s):

A baseline assessment is performed during the screening visit. It is then assessed again at 1 year, and 3 years of follow-up after the last study treatment (LST).

Operationalisation:

The QoL (QLQ-C30) outcome variable is operationalised through the QLQ-C30 questionnaire. The answers are transformed into 3 separate scores, which can all range from 0 to 100. These scores are: Global health status/QoL, Functional scales, Symptom scales. Where a higher score on the first 2 scales indicate a better health state and QoL, while on the symptom scale a higher score indicates a higher level of disturbance by symptoms.

Analysis metric:

The QLQ-C30 Global health status absolute values will be used for analysis.

Method of aggregation:

The mean, median, SD and IQR will be reported. For comparisons and estimands, please refer to the SAP.

Time point(s):

A baseline assessment is performed during the screening visit. It is then assessed again at 1 year, and 3 years of follow-up after the last study treatment (LST).

Operationalisation:

The QoL (QLQ-C30) outcome variable is operationalised through the QLQ-C30 questionnaire. The answers are transformed into 3 separate scores, which can all range from 0 to 100. These scores are: Global health status/QoL, Functional scales, Symptom scales. Where a higher score on the first 2 scales indicate a better health state and QoL, while on the symptom scale a higher score indicates a higher level of disturbance by symptoms.

Analysis metric:

The QLQ-C30 Functional scales score absolute values will be used for analysis.

Method of aggregation:

The mean, median, SD and IQR will be reported. For comparisons and estimands, please refer to the SAP.

Time point(s):

A baseline assessment is performed during the screening visit. It is then assessed again at 1 year, and 3 years of follow-up after the last study treatment (LST).

Operationalisation:

The QoL (QLQ-C30) outcome variable is operationalised through the QLQ-C30 questionnaire. The answers are transformed into 3 separate scores, which can all range from 0 to 100. These scores are: Global health status/QoL, Functional scales, Symptom scales. Where a higher score on the first 2 scales indicate a better health state and QoL, while on the symptom scale a higher score indicates a higher level of disturbance by symptoms.

Analysis metric:

The QLQ-C30 Symptom scales score absolute values will be used for analysis.

Method of aggregation:

The mean, median, SD and IQR will be reported. For comparisons and estimands, please refer to the SAP.

Time point(s):

A baseline assessment is performed during the screening visit. It is then assessed again at 1 year, and 3 years of follow-up after the last study treatment (LST).

Operationalisation:

The QoL (QLQ-BR23) outcome variable is operationalised through the QLQ-BR23 questionnaire. The answers are transformed into 2 separate scores, which can both range from 0 to 100. These scores are: Functional scales; Symptom scales. Where a higher score on the functional scale indicate a better health state and QoL, while on the symptom scale a higher score indicates a higher level of disturbance by symptoms.

Analysis metric:

The QLQ-BR23 Functional scales score absolute values will be used for analysis.

Method of aggregation:

The mean, median, SD and IQR will be reported. For comparisons and estimands, please refer to the SAP.

Time point(s):

A baseline assessment is performed during the screening visit. It is then assessed again at 1 year, and 3 years of follow-up after the last study treatment (LST).

Operationalisation:

The QoL (QLQ-BR23) outcome variable is operationalised through the QLQ-BR23 questionnaire. The answers are transformed into 2 separate scores, which can both range from 0 to 100. These scores are: Functional scales; Symptom scales. Where a higher score on the functional scale indicate a better health state and QoL, while on the symptom scale a higher score indicates a higher level of disturbance by symptoms.

Analysis metric:

The QLQ-BR23 Symptom scales score absolute values will be used for analysis.

Method of aggregation:

The mean, median, SD and IQR will be reported. For comparisons and estimands, please refer to the SAP.

Time point(s):

A baseline assessment is performed during the screening visit. It is then assessed again at 1 year, and 3 years of follow-up after the last study treatment (LST).

Operationalisation:

During the 1Y and 3Y Follow-Up study visits the overall and tumour bed fibrosis will be scored according to the 'National Cancer Institute Common Terminology Criteria for Adverse Events' (NCI-CTCAE) v5.0 AE reporting system. This scoring will be performed by either the principal- or sub-investigators.

Analysis metric:

Tabulation of absence vs. presence and severity of fibrosis will be used for analysis.

Method of aggregation:

The proportion of participants experiencing no 'Fibrosis' vs. any 'Fibrosis' will be used. As well as the frequency of all grades, including 0 (absence).

For comparisons and estimands, please refer to the SAP.

Time point(s):

Fibrosis will be assessed and recorded during the 1YFU and 3YFU study visits.

Operationalisation:

During the 1Y and 3Y Follow-Up study visits the occurrence of 'Breast pain' will be scored according to the 'National Cancer Institute Common Terminology Criteria for Adverse Events' (NCI-CTCAE) v5.0 AE reporting system. This scoring will be performed by either the principal- or sub-investigators.

Analysis metric:

Tabulation of absence vs. presence and severity of 'Breast pain' will be used for analysis.

Method of aggregation:

The proportion of participants experiencing no 'Breast pain' vs. any 'Breast pain' will be used. As well as the frequency of all grades, including 0 (absence).

For comparisons and estimands, please refer to the SAP.

Time point(s):

Fibrosis will be assessed and recorded during the 1YFU and 3YFU study visits.

Operationalisation:

The oncological survival and TTE data are operationalised as time-to-event intervals for the events of interest listed below.

The 'OS' TTE/survival outcome will be reported according to the 2015 DATECAN consensus, including time-to-event data for the following events of interest: 'Death from any cause'.

Analysis metric:

The TTE data is registered in days from randomisation (Rz). Censoring will be used for participants without events at the end of their follow up.

Method of aggregation:

KM-estimates, as well as proportions free from events at follow-up visit timepoints will be reported. The non-aggregated data will be used for survival analysis. For comparisons and estimands, please refer to the SAP.

Operationalisation:

The oncological survival and TTE data are operationalised as time-to-event intervals for the events of interest listed below.

The 'BCSS' TTE/survival outcome will be reported according to the 2015 DATECAN consensus, including time-to-event data for the following events of interest: 'Death from breast cancer'.

Analysis metric:

The TTE data is registered in days from randomisation (Rz). Censoring will be used for participants without events at the end of their follow up.

Method of aggregation:

KM-estimates, as well as proportions free from events at follow-up visit timepoints will be reported. The non-aggregated data will be used for survival analysis. For comparisons and estimands, please refer to the SAP.

Operationalisation:

The oncological survival and TTE data are operationalised as time-to-event intervals for the events of interest listed below.

The 'RFS' TTE/survival outcome will be reported according to the 2015 DATECAN consensus, including time-to-event data for the following events of interest: 'Death from any cause', 'Invasive ipsilateral breast tumor recurrence', 'Local Invasive recurrence', 'Regional Invasive recurrence', 'occurrence of metastases', or 'Ipsilateral DCIS'.

Analysis metric:

The TTE data is registered in days from randomisation (Rz). Censoring will be used for participants without events at the end of their follow up.

Method of aggregation:

KM-estimates, as well as proportions free from events at follow-up visit timepoints will be reported. The non-aggregated data will be used for survival analysis. For comparisons and estimands, please refer to the SAP.

Operationalisation:

The oncological survival and TTE data are operationalised as time-to-event intervals for the events of interest listed below.

The 'L-RFS' TTE/survival outcome will be reported according to the 2015 DATECAN consensus, including time-to-event data for the following events of interest: 'Death from any cause', 'Invasive ipsilateral breast tumor recurrence', 'Local Invasive recurrence', 'Regional Invasive recurrence', or 'Ipsilateral DCIS'.

Analysis metric:

The TTE data is registered in days from randomisation (Rz). Censoring will be used for participants without events at the end of their follow up.

Method of aggregation:

KM-estimates, as well as proportions free from events at follow-up visit timepoints will be reported. The non-aggregated data will be used for survival analysis. For comparisons and estimands, please refer to the SAP.

Operationalisation:

The oncological survival and TTE data are operationalised as time-to-event intervals for the events of interest listed below.

The 'D-RFS' TTE/survival outcome will be reported according to the 2015 DATECAN consensus, including time-to-event data for the following events of interest: 'Death from any cause', or 'occurrence of metastases'.

Analysis metric:

The TTE data is registered in days from randomisation (Rz). Censoring will be used for participants without events at the end of their follow up.

Method of aggregation:

KM-estimates, as well as proportions free from events at follow-up visit timepoints will be reported. The non-aggregated data will be used for survival analysis. For comparisons and estimands, please refer to the SAP.