A systematic review and meta‐analysis of predictive and prognostic models for outcome prediction using positron emission tomography radiomics in head and neck squamous cell carcinoma patients

Abstract

BackgroundPositron emission tomography (PET) images of head and neck squamous cell carcinoma (HNSCC) patients can assess the functional and biochemical processes at cellular levels. Therefore, PET radiomics‐based prediction and prognostic models have the potentials to understand tumour heterogeneity and assist clinicians with diagnosis, prognosis and management of the disease. We conducted a systematic review of published modelling information to evaluate the usefulness of PET radiomics in the prediction and prognosis of HNSCC patients.MethodsWe searched bibliographic databases (MEDLINE, Embase, Web of Science) from 2010 to 2021 and considered 31 studies with pre‐defined inclusion criteria. We followed the CHARMS checklist for data extraction and performed quality assessment using the PROBAST tool. We conducted a meta‐analysis to estimate the accuracy of the prediction and prognostic models using the diagnostic odds ratio (DOR) and average C‐statistic, respectively.ResultsManual segmentation method followed by 40% of the maximum standardised uptake value (SUVmax) thresholding is a commonly used approach. The area under the receiver operating curves of externally validated prediction models ranged between 0.60–0.87, 0.65–0.86 and 0.62–0.75 for overall survival, distant metastasis and recurrence, respectively. Most studies highlighted an overall high risk of bias (outcome definition, statistical methodologies and external validation of models) and high unclear concern in terms of applicability. The meta‐analysis showed the estimated pooled DOR of 6.75 (95% CI: 4.45, 10.23) for prediction models and the C‐statistic of 0.71 (95% CI: 0.67, 0.74) for prognostic models.ConclusionsBoth prediction and prognostic models using clinical variables and PET radiomics demonstrated reliable accuracy for detecting adverse outcomes in HNSCC, suggesting the prospect of PET radiomics in clinical settings for diagnosis, prognosis and management of HNSCC patients. Future studies of prediction and prognostic models should emphasise the quality of reporting, external model validation, generalisability to real clinical scenarios and enhanced reproducibility of results.

Methodological evaluation of original articles on radiomics and machine learning for outcome prediction based on positron emission tomography (PET)

“…The median number of criteria with an "adequate" rating per article was 12.5 (IQR, 9-14; range, [4][5][6][7][8][9][10][11][12][13][14][15][16][17]. Two articles [12,42] were rated "adequate" in all 17 criteria, whereas 22 articles (22 %) had an "adequate" rating in less than half of the items (≤ 8 of 17 items).…”

“…These metrics are most commonly derived from manually or semi-automatically delineated regions of interest. In recent years, radiomics and machine learning-based prediction models have been increasingly employed to enhance the prognostic or predictive value of PET imaging by leveraging textural information and patterns that are not directly accessible to human readers [4,5]. However, the increasing complexity and feature number of such approaches, compared to the sparsity of manually derived image features, brings with it a higher risk of obtaining results that are either biased or not reproducible.…”

MRI radiomics in head and neck cancer from reproducibility to combined approaches

“…Overall, although the literature includes several meta-analysis studies on radiomics 39 – 41 , to the best of the authors’ knowledge, the present study represents the initial endeavor to (1) replicate published radiomic signatures on an external dataset, offering a potential method to address the insufficient reporting, (2) provide a detailed characterization of the reproduced radiomic signatures, and (3) propose combined approaches to enhance the prognostic performance. The proposed study yielded to key findings.…”

Developments and future prospects of personalized medicine in head and neck squamous cell carcinoma diagnoses and treatments

“…Despite advancements in therapeutic drugs, the challenge persists as many patients are identified in late stages due to the specific growth site of HNSCC. 207 , 208 Consequently, the prognosis remains weak, with the five‐year survival rate varying for hypopharyngeal cancer from 25% and upto 80% for NPC. Consequently, there is a strong interest among scientists to improve the accuracy of predicting LR, LNM, and even DM in HNSCC, along with enhancing the prediction of patients' survival rates.…”