Predicting the Risk of Disease Recurrence and Death Following Curative-intent Radiotherapy for Non-small Cell Lung Cancer: The Development and Validation of Two Scoring Systems From a Large Multicentre UK Cohort

Abstract

Introduction:There is a paucity of evidence on which to produce recommendations on neither the clinical nor the imaging follow-up of lung cancer patients after curative intent radiotherapy. In the 2019 NICE lung cancer guidelines further research into risk stratification models to inform follow up protocols was recommended. Methods: A retrospective study of consecutive patients undergoing curative-intent radiotherapy for NSCLC from 01/10/2014 to 01/10/2016 across nine UK trusts. Twenty two demographic, clinical and treatment-related variables were collected and multivariable logistic regression used to develop and validate two risk stratification models to determine the risk of disease recurrence and death. Results: 898 patients were included in the study. Mean age was 72yrs, 63% (562/898) were good PS (0-1) and 43% (388/898), 15% (134/898) & 42% (376/898) were clinical stage I, II and III respectively. 36% (322/898) suffered disease recurrence and 41% (369/898) in the first two years following radiotherapy. The ASSENT score (Age, performance Status, Smoking status, staging EBUS, N-stage, Tstage) was developed that stratifies the risk for disease recurrence within 2 years with an AUROC for the total score of 0.712 (0.671-0.753) and 0.72 (0.65-0.789) in the derivation and validation set respectively. The STEPS score (Sex, Performance status, staging EBUS, T-stage, N-Stage) was developed that stratifies the risk of death within 2 years with an AUROC for total score of 0.625 (0.581-0.669) and 0.607 (0.53-0.684) in the derivation and validation set respectively.Conclusions: These validated risk stratification models could be used to inform follow-up protocols after curative intent radiotherapy for lung cancer. The modest performance highlights the need for more advanced risk prediction tools.

A comparison of machine learning methods for predicting recurrence and death after curative-intent radiotherapy for non-small cell lung cancer: Development and validation of multivariable clinical prediction models

“… 2 Recurrence is reported in up to 36% of patients receiving curative-intent treatment for NSCLC. 3 Surveillance is recommended across international guidelines for NSCLC patients treated with curative-intent radiotherapy. 3 Surveillance ensures on-going patient support, management of co-morbidities and cancer treatment-related side effects as well as detection of recurrence of the treated cancer or second (metachronous) primary cancers.…”

“… 3 Surveillance is recommended across international guidelines for NSCLC patients treated with curative-intent radiotherapy. 3 Surveillance ensures on-going patient support, management of co-morbidities and cancer treatment-related side effects as well as detection of recurrence of the treated cancer or second (metachronous) primary cancers. Curative treatment following local recurrence results in 5-year survival rates of 15%.…”

“…International guidance highlights a lack of high-quality evidence to formulate specific recommendations on the nature and frequency of follow-up after radiotherapy. 3 Most guidance is based on expert consensus and advises follow-up with clinical review and contrast-enhanced chest CT at 3–6 monthly intervals for 2–3 years. Without an agreed framework, follow-up differs according to local policy or even individual clinician opinion.…”

“…Previous studies using clinical data for outcome prediction following radiotherapy for NSCLC have explored cox proportional-hazards models and support vector machines, and have generally focused on OS, achieving validation or external test set AUCs between 0·61 and 0·69 and C-Index scores between 0·58 and 0·69. 3 , 5 , 6 , 7 , 8 It is possible that more advanced machine learning techniques can better handle large clinical datasets leading to superior performance. Comparison of the performance of several machine learning models has been undertaken for systemic anti-cancer therapy for advanced NSCLC, 9 but is lacking in the radical radiotherapy cohort.…”

Risk factors for radiation-induced lung injury in patients with advanced non-small cell lung cancer: implication for treatment strategies

“…At the same time, if the patient has diabetes, smoking, COPD, concurrent chemotherapy, and radiotherapy, the total dose should be optimized and preventive measures should be taken to reduce the occurrence of RILI [41]. Previous studies [42,43] have shown that complications such as chronic obstructive pulmonary disease and ventilatory dysfunction caused by smoking can increase the risk of lung injury. It is recommended that the lung diffusion function be tested before radiotherapy to evaluate the status of lung function.…”