Free cookie consent management tool by TermsFeed

ESGAR

Education

life long learning

Journal Watch - September 2021 (1)

MRI radiomics features of mesorectal fat can predict response to neoadjuvant chemoradiation therapy and tumor recurrence in patients with locally advanced rectal cancer

Jayaprakasam VS, Paroder V, Gibbs P, Bajwa R, Gangai N, Sosa RE, Petkovska I, Golia Pernicka JS, Fuqua JL 3rd, Bates DDB, Weiser MR, Cercek A, Gollub MJ.

Eur Radiol. 2021 Jul 29. doi: 10.1007/s00330-021-08144-w. Epub ahead of print.

 

Colorectal cancer is the third most common cancer worldwide and the second cause of oncologic-related mortality. Rectal cancer accounts for ~30% of the total cases of colorectal malignancies (1). The standard-of-care treatment for patients with locally advanced rectal cancer (LARC) is neoadjuvant chemotherapy (nCRT) followed by total mesorectal excision (TME) (2). However, the individual response is very heterogeneous, ranging from a pathological complete response to no tumor regression or even disease progression (3). Recently, radiomics has gained increasing interest, more and more papers investigating the efficacy of radiomics features as biomarkers for lesion characterization and tumor prognosis, including for rectal cancer (4,5).

In this new study by Jaypar and colleagues, the focus is on the radiomics features extracted not from the rectal tumor, but from the mesorectal fat. The authors aimed to assess the potential role of radiomics analysis of mesorectal fat for the prediction of clinical outcomes of LARC.

This retrospective study included 236 patients with LARC who underwent total nCRT and TME. The following clinical outcomes were evaluated: pathologic complete response (pCR), local and distant recurrence, clinical (cT) and post-treatment (ypT) T categories, clinical (cN) and post-treatment (ypN) N categories, and clinical (cTNM) and post-treatment (ypTNM) TNM stages. Most patients had a cT category <4 (n=201), cN category>0 (n=213), post-treatment ypT category >0 (n=190), were post-treatment node-negative patients (n=176) and without local or distant recurrence (n=164, n=171 respectively). 101 radiomics features were extracted from the unfiltered, T2-weighted axial images of the baseline rectal MRI scans. Mesorectal fat segmentation was performed individually by three radiologists, excluding other structures such as tumor deposits, lymph nodes and extramural vascular invasion. Only features with a good intraclass coefficient (ICC>0.6) were included for further analysis (81 features).

Mann-Whitney test was firstly employed to select the significantly different radiomics parameters among each clinical outcome. Afterwards, using support vector machine algorithms and 5-fold cross validation, the authors developed different predictive models for each outcome. ROC curve analysis was conducted and area under the curve (AUC), sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were calculated to evaluate the performance of each model. The best performing model was the one for distinguishing between pCR and non-responders (AUC=0.89, Se=78.0%, Sp=85.1%). Also, radiomics models significantly distinguished patients with local and distant recurrence from those with no recurrences (AUC=0.79 and 0.87, respectively). The resulting AUCs for discriminating between patients with cT4 category/those with <cT4 category and between post-treatment node-negative patients/node-positive patients were 0.80 and 0.74 respectively. The radiomics model for predicting the post-treatment ypT category achieved an AUC=0.86. However, the radiomics features did not differ significantly between cN<0 and cN>0 patients or between cTNM stage 2 and stage 3 patients, therefore no models were built for these outcomes.

The novelty of this research consists of shifting the focus of radiomics analysis from the tumour itself to the peritumoral environment in predicting cancer outcome. This issue has been investigated in several malignancies such as breast or lung cancer (6,7). However, only one recent study by Shaish et al included features obtained from the peritumoral region in the development of predictive models for assessing tumor response and prognosis of rectal cancer (8). Similar with the current study, their research has reported promising results, achieving an AUC=0.80 for predicting pCR using only features derived from mesorectal compartment.

Evaluation of the mesorectal fat may offer valuable prognostic information since the growth and dissemination of rectal cancer is closely related with the changes of the nearby adipocytes. Recent papers have showed that metabolic alterations in cancer-associated adipocytes can lead to secretion of various growth factors and remodeling factors that can interfere with cancer development and even therapeutic response (9,10). Therefore, assessment of peritumoral fatty environment may represent a new area of research, not only for rectal cancer but also for other types of abdominal/gastrointestinal tumors.

This paper has a few limitations, namely its retrospective design and the limited sample size. Due to the latter one, the data were not split into test and training sets, using a cross-validation approach instead. Still, this is the largest study which investigated the potential of mesorectal fat radiomics analysis for prediction of cancer outcome and it may represent a pathfinder for future research of peritumoral environment.

 

References:

 

  1. Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer Statistics, 2021. CA Cancer J Clin. 2021 Jan;71(1):7-33.
  2. Benson AB, Venook AP, Al-Hawary MM, Cederquist L., Chen YJ, Ciombor KK, Cohen S, Cooper HS, Deming D, Engstrom PF et al. Rectal Cancer, Version 2.2018, NCCN clinical practice guidelines in oncology. J. Natl. Compr. Cancer Netw. 2018, 16, 874–901
  3. Kong, JC, Guerra GR, Warrier SK, Lynch, AC, Michael, M, Ngan, SY, Phillips, W, Ramsay, G, Heriot, AG. Prognostic value of tumour regression grade in locally advanced rectal cancer: A systematic review and meta-analysis. Colorectal Dis. 2018, 20, 574–585
  4. Cui Y, Yang X, Shi Z, Yang Z, Du X, Zhao Z, Cheng, X. Radiomics Analysis of Multiparametric MRI for Prediction of Pathological Complete Response to Neoadjuvant Chemoradiotherapy in Locally Advanced Rectal Cancer. Eur. Radiol. 2019, 29, 1211–1220.
  5. Petresc B, Lebovici A, Caraiani C, Feier DS, Graur F, Buruian MM. Pre-Treatment T2-WI Based Radiomics Features for Prediction of Locally Advanced Rectal Cancer Non-Response to Neoadjuvant Chemoradiotherapy: A Preliminary Study. Cancers (Basel). 2020 Jul 14;12(7):1894.
  6. Braman NM, Etesami M, Prasanna P, Dubchuk C, Gilmore H, Tiwari P, Plecha D, Madabhushi A. Intratumoral and peritumoral radiomics for the pretreatment prediction of pathological complete response to neoadjuvant chemotherapy based on breast DCE-MRI. Breast Cancer Res. 2017 May 18;19(1):57.
  7. Pérez-Morales J, Tunali I, Stringfield O, Eschrich SA, Balagurunathan Y, Gillies RJ, Schabath MB. Peritumoral and intratumoral radiomic features predict survival outcomes among patients diagnosed in lung cancer screening. Sci Rep. 2020 Jun 29;10(1):10528.
  8. Shaish H, Aukerman A, Vanguri R, Spinelli A, Armenta P, Jambawalikar S, Makkar J, Bentley-Hibbert S, Del Portillo A, Kiran R, Monti L, Bonifacio C, Kirienko M, Gardner KL, Schwartz L, Keller D. Radiomics of MRI for pretreatment prediction of pathologic complete response, tumor regression grade, and neoadjuvant rectal score in patients with locally advanced rectal cancer undergoing neoadjuvant chemoradiation: an international multicenter study. Eur Radiol. 2020 Nov;30(11):6263-6273.
  9. Duong MN, Geneste A, Fallone F, Li X, Dumontet C, Muller C. The fat and the bad: Mature adipocytes, key actors in tumor progression and resistance. Oncotarget. 2017 May 20;8(34):57622-57641.
  10. Jiramongkol Y, Lam EW. Multifaceted Oncogenic Role of Adipocytes in the Tumour Microenvironment. Adv Exp Med Biol. 2020;1219:125-142.

    Dr. Bianca Boca (Petresc) is a fifth-year radiology resident at the County Clinical Emergency Hospital in Cluj-Napoca, Romania. Her main areas of focus include abdominal and urogenital imaging, especially rectal and liver imaging. She has been developing a particular interest in using software for texture analysis and since 2019 she is a PhD student at “George Emil Palade” University of Medicine, Pharmacy, Science and Technology of Târgu Mureș, investigating the role of radiomics in pelvic cancers. Passionate about research, Bianca has actively participated in previous ESGAR/ECR events and has published several papers as an author/co-author in international journals.

    Comments may be sent to