Original paperCharacterization of cervical lymph-nodes using a multi-parametric and multi-modal approach for an early prediction of tumor response to chemo-radiotherapy
Introduction
Treatment of locally advanced Head-and-Neck Squamous Cell Carcinoma (HNSCC) is a challenging task, since it affects the overall quality of life of patients, and, despite chemo-radiotherapy (CRT) has been accepted as a standard management, the 5 year survival rate of these patients remains below 50% [1]. In this context, a timely identification of patients at risk of residual malignant lymph nodes (LNs) can help in choosing the best treatment [2]. However, the early prediction of treatment response presents many challenges and it is currently a hot topic in this research area, since reproducible and robust findings have not been established yet [3].
Recently, many studies have reported image-based criteria for treatment assessment, considering Computed Tomography (CT), morphological or functional Magnetic Resonance Imaging (MRI) [3], [4], [5], [6], [7], [8], [9]. The morphological criteria principally regard volume regression [5], [7], [9], the presence of focal abnormalities [3] and nodal density [8]. However, very heterogeneous results in the prediction performance using these criteria were reported in literature [10].
Great attention was given to Diffusion-weighted MRI (DW-MRI) [11] in the evaluation of diffusion properties of the primary tumor and malignant nodes, related to the treatment response [12]. In fact, it was reported that primary tumors and metastatic nodes characterized by lower pre-treatment values of apparent diffusion coefficient (ADC) can better respond to the CRT [13], [14], [15]. Moreover, ADC changes evaluated both during and early after CRT were found predictive of treatment response at the primary and nodal site [6], [10], [16].
Very few works [17], [18] tried to perform a multi-parametric analysis, but they were limited to a combination of ADC and morphological features or ADC and PET findings. Moreover, morphological evaluations were performed using qualitative findings or very simple quantitative approaches, able to characterize nodes density or shape [3], [9]. To our knowledge, there are no attempts to predict treatment response using advanced image analysis techniques, such as texture analysis, on anatomical images which can provide quantitative information about structural properties of the tissue, related to the spatial pattern. Only a very recent work proposed textural features applied on parametric maps derived on Dynamic Contrast Enhanced (DCE)-MRI for the prediction of tumor response [19].
Aim of this work was thus to investigate the potential of a multi-modal characterization descriptive of morphological, structural and functional properties at baseline and at mid-CRT, based on advanced image processing methods, for the early prediction of LNs response to CRT in patients with HNSCC. In particular, shape-related indices were calculated on T2w-MRI, textural features were extracted from T2w-MRI and planning CT, and ADC values from DW-MRI.
Section snippets
Patient population
From October 2011 to March 2014, a total of 30 patients with histologically proven HNSCC were included retrospectively in the present study. This retrospective analysis was performed on an existing data-set of a prospective study conducted at Regina Elena National Cancer Institute and previously approved by the ethics committee of the hospital. The primary end point of the study, currently closed, was to investigate the ability of DWI to predict the treatment response of HNSCC patients after
Patient population
Nineteen LNs (63.3%) showed RC after a median follow-up time of 26.6 months (range, 7.3, 50.2). Eleven LNs (36.7%) showed RF at a median time of 4.6 months (range, 1.5, 19.5), four of which had a persistence of the lesion at the first MR scan performed 8 weeks after CRT. Three of the 11 patients with RF developed also a primary tumor recurrence. In two patients, PET-CT acquired 12 weeks after the end of CRT documented a recurrence in a LN different from the largest one. The patients with RC and RF
Discussion
In this paper, a multi-modal and multi-parametric characterization of cervical lymph nodes was proposed to predict the specific response to chemo-radiotherapy in patients treated for HNSCC, using quantitative image-based features calculated before and at mid treatment.
The patient population analysis demonstrated that the two groups were homogeneous regarding baseline characteristics. On the contrary, the image-based characterization showed significant differences, especially in terms of pre-RT
Conclusion
The present work suggested that a multi-modal approach, aimed at characterizing cervical nodes with different textural and quantitative features, can early predict treatment response in patients with HNSCC with good accuracy. When textural features estimated on T2w-MRI were combined with basal ADC, the prediction power increased. Performance equivalent to ADC, which can be considered in this context as the gold-standard biomarker, can be obtained by the combination of textural features, both
Funding sources
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References (34)
- et al.
Lymph node positive head and neck carcinoma after curative radiochemotherapy: a long lasting debate on elective post-therapeutic neck dissections comes to a conclusion
Cancer Radiothérapie J La Société Fr Radiothérapie Oncol
(2013) - et al.
Prediction of treatment response in head and neck carcinomas using IVIM-DWI: evaluation of lymph node metastasis
Eur J Radiol
(2014) - et al.
Characteristics and kinetics of cervical lymph node regression after radiation therapy for human papillomavirus-associated oropharyngeal carcinoma: quantitative image analysis of post-radiotherapy response
Oral Oncol
(2015) - et al.
Detection of head and neck squamous cell carcinoma with diffusion weighted MRI after (chemo)radiotherapy: correlation between radiologic and histopathologic findings
Int J Radiat Oncol Biol Phys
(2007) - et al.
Pretreatment apparent diffusion coefficient of the primary lesion correlates with local failure in head-and-neck cancer treated with chemoradiotherapy or radiotherapy
Int J Radiat Oncol Biol Phys
(2011) - et al.
Simultaneous 18F-FDG-PET/MRI: associations between diffusion, glucose metabolism and histopathological parameters in patients with head and neck squamous cell carcinoma
Oral Oncol
(2016) - et al.
3D Slicer as an image computing platform for the Quantitative Imaging Network
Magn Reson Imaging
(2012) - et al.
Integrating pretreatment diffusion weighted MRI into a multivariable prognostic model for head and neck squamous cell carcinoma
Radiother Oncol
(2014) - et al.
Texture analysis for the assessment of structural changes in parotid glands induced by radiotherapy
Radiother Oncol
(2013) - et al.
Radiation-induced damage to microstructure of parotid gland: evaluation using high-resolution magnetic resonance imaging
Int J Radiat Oncol Biol Phys
(2010)
Whole-body-MR imaging including DWIBS in the work-up of patients with head and neck squamous cell carcinoma: A feasibility study
Eur J Radiol
Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck
N Engl J Med
Cervical nodal metastases from head and neck squamous cell carcinoma: MRI criteria for treatment assessment
Head Neck
Image guided volumetric response during chemoradiotherapy for head and neck squamous cell carcinoma as a predictor of disease outcomes
Am J Otolaryngol
Intravoxel incoherent motion imaging kinetics during chemoradiotherapy for human papillomavirus-associated squamous cell carcinoma of the oropharynx: preliminary results from a prospective pilot study
NMR Biomed
Volumetric change of human papillomavirus-related neck lymph nodes before, during, and shortly after intensity-modulated radiation therapy
Head Neck
Early volume variation of positive lymph nodes assessed by in-room mega voltage CT images predicts risk of loco-regional relapses in head and neck cancer patients treated with intensity-modulated radiotherapy
Acta Oncol
Cited by (23)
The Use of MR-Guided Radiation Therapy for Head and Neck Cancer and Recommended Reporting Guidance
2024, Seminars in Radiation OncologyQuantitative magnetic resonance imaging on hybrid magnetic resonance linear accelerators: Perspective on technical and clinical validation
2020, Physics and Imaging in Radiation OncologyCitation Excerpt :There is growing evidence that diffusion weighted (DW) MRI may provide prognostic information to predict RT treatment outcome in head and neck cancer (HNC) [7–11]. Also for other tumor sites, such as brain tumors [12,13], cervical cancer [14] or rectal cancer [15], the prognostic value of DW MRI for RT outcome prognosis has been shown. The majority of these studies correlated functional biomarkers derived from functional MRI before the start of RT with observed outcome following therapy.
PETER PHAN: An MRI phantom for the optimisation of radiomic studies of the female pelvis
2020, Physica MedicaCitation Excerpt :Moreover, the lack of a standard intensity scale for MR images necessitates the use of additional pre-processing, such as intensity normalisation and bias field correction [7]. Early clinical demonstrations of radiomics in MRI have been seen in prostate [8] and breast cancer [9], as well as glioblastoma multiforme [10] and head-and-neck squamous cell carcinoma [11]. Promising results have also been seen in female pelvic tumours, where preoperative radiomic features were seen to be predictive of lymph-vascular space invasion [12] and of lymph node metastasis [13] in cervical cancer patients.
Functional imaging early during (chemo)radiotherapy for response prediction in head and neck squamous cell carcinoma; a systematic review
2019, Oral OncologyCitation Excerpt :An ADCmean increase after 1 week of treatment was reported higher in LRC than in LRF patients [47]. At 2–3 weeks two studies did not find significant differences in ΔADCmean between LRC and LRF patients [52,53], whereas three other studies showed an overall trend towards a higher ADCmean increase at 3 weeks intratreatment in LRC (22%[45], >25%[54], 100% increase [56]) compared with LRF (7%[45], not specified [54], 38% increase [56]). An ADCmean increase of 25% at 4 weeks was reported in LRC [55].
- 1
E.S. and S.M. equally contributed to the work.