Early weight loss during radiotherapy or chemoradiotherapy (CRT) is well-known to be associated with decreased progression-free survival (PFS) and overall survival (OS).1,2 Particularly in patients receiving concurrent CRT, weight loss is a frequent, often expected, occurrence.
It is suggested that prevention of weight loss with early nutritional intervention and physical activity may improve the patient’s compliance with treatment, decrease side effects, and improve PFS and OS.3 However, high-level evidence is lacking. It is also likely that not all weight loss can be prevented with nutritional intervention, as the mechanisms may be diverse, including such causes as decreased food intake due to esophagitis as well as systemic inflammation. The former will likely benefit from dietary measures, and the latter less so.
Nevertheless, identifying patients who are at risk for early weight loss during CRT could result in targeted interventions for only those patients who are most likely to need it, thus saving resources. In the study of Han et al.,4 a clinical decision support system was created on the basis of radiomics, dosiomics extracted from CT scans, and 3D dose-volume parameters. The positive predictive value (PPV) was 0.35 to 0.40, and the negative predictive value (NPV) 0.76 to 0.82. These were better than the predictions of physicians, whose performance was close to randomness.
A major caveat of this interesting prospective study, aside from its small size with 37 patients, is that physicians had to predict severe weight loss at the time of patient intake without knowledge of the dose distribution, for instance, to the esophagus. It is possible this insight might have increased the accuracy of clinicians’ predictions. Physicians tend to make predictions with a high NPV and a low PPV in an effort to avoid overlooking a patient at risk for developing side effects. The study did not assess sarcopenia at the time of diagnosis, which is a well-known prognostic factor for weight loss and poor PFS and OS.5 It is therefore unclear whether the current model predicts only weight loss that is due to decreased intake because of mechanical reasons such as esophagitis, or whether biochemical mechanisms were also considered.
Regardless, the Han study adds to the body of evidence demonstrating that emerging techniques such as radiomics and dosiomics, together with artificial intelligence, can improve prognostic and predictive models and aid clinical decision making. However, to achieve clinical implementation, rigorous testing in high-quality studies with sufficient power is needed.6 At the same time, “explainable artificial intelligence” may be needed to make the algorithm reliable and versatile enough for clinicians to understand the boundaries for its use, similar to inclusion and exclusion criteria in regular clinical trials. If that is realized, models such as that of Han et al. will become as common as a stethoscope in daily clinical care.
References:
- 1. Sanders KJ, Hendriks LE, Troost EG, et al. Early Weight Loss during Chemoradiotherapy Has a Detrimental Impact on Outcome in NSCLC. J Thorac Oncol. 2016;11(6):873-879.
- 2. Willemsen ACH, Degens JHRJ, Baijens LWJ, et al. Early Loss of Fat Mass During Chemoradiotherapy Predicts Overall Survival in Locally Advanced Squamous Cell Carcinoma of the Lung, but Not in Locally Advanced Squamous Cell Carcinoma of the Head and Neck. Front Nutr. 2020;7:600612.
- 3. De Ruysscher D, Faivre-Finn C, Nackaerts K, et al. Recommendation for supportive care in patients receiving concurrent chemotherapy and radiotherapy for lung cancer. Ann Oncol. 2020;31(1):41-49.
- 4. Han P, Lee SH, Noro K, et al. Improving Early Identification of Significant Weight Loss Using Clinical Decision Support System in Lung Cancer Radiation Therapy. JCO Clin Cancer Inform. 2021;5(2021):944-952.
- 5. Burtin C, Bezuidenhout J, Sanders KJC, et al. Handgrip weakness, low fat-free mass, and overall survival in non-small cell lung cancer treated with curative-intent radiotherapy. J Cachexia Sarcopenia Muscle. 2020;11(2):424-431.
- 6. O’Connor JP, Aboagye EO, Adams JE, et al. Imaging biomarker roadmap for cancer studies. Nat Rev Clin Oncol. 2017;14(3):169-186.
