Tumour response in non-small-cell lung cancer patients treated with chemoradiotherapy - Can spectral CT predict recurrence?

Abstract

Tumour response in lung cancer treatment is monitored by measuring lesion size in computed tomography (CT). Spectral CT (SCT) offers additional information on tumour tissue besides morphology. We evaluated SCT iodine content (IC) and performed spectral slope analysis to assess the response of non-small-cell lung cancer (NSCLC) to chemoradiotherapy (CRT). Eighty-three patients with advanced NSCLC treated by CRT prospectively underwent single-phase, contrast-enhanced SCT. Evaluation of all patients included treatment response (RECIST 1.1), quantitative measurements as well as SCT IC determination and spectral slope analysis in NSCLC primaries. Measurements were performed at the maximum cross-diameter of tumours and in areas with high iodine values (hotspot analysis). Iodine difference (ΔIC) was calculated. Secondary outcome parameters were IC and spectral slopes in mediastinal lymph nodes (n=61). Twenty-four patients (29%) showed complete remission after CRT. Thirty-four patients (41%) had stable disease (SDSCT ) or partial regression (PRSCT ). Progressive disease (PDSCT ) was seen in 25 patients (30%). Hotspot analysis showed significantly higher iodine values in PDSCT than in SDSCT /PRSCT (P<0.001). Ten patients (12%) with initially stable disease in SCT showed progressive disease during follow-up for up to 18months (PDFU ). These patients also had significantly higher hotspot iodine values and ΔIC in the initial scan compared to patients with SD throughout the follow-up period (SDFU ) (29%) (P<0.001). Enlarged lymph nodes showed significantly lower iodine content and a lower spectral slope pitch than normal-sized nodes (P=0.003 to 0.029). Spectral CT-derived iodine content of NSCLC following CRT may help in predicting recurrence. Hotspot analysis and iodine heterogeneity allow the identification of residual vascularisation as an indicator of vital tumour tissue, indicating that IC might be a suitable imaging biomarker for predicting tumour progression. Iodine content and spectral slope analysis might also help in identifying metastatic lymph nodes.