Tumor-stroma ratio (TSR) in non-small cell lung cancer (NSCLC) patients after lung resection is a prognostic factor for survival

Xi, Ke Xing; Wen, Ying Sheng; Zhu, Chong Mei; Yu, Xiang; Qin, Rong; Zhang, Xuewen; Lin, Yong; Rong, Tie Hua; Wang, Weidong; Chen, Yong Qiang; Zhang, Lan Jun

doi:10.21037/jtd.2017.09.29

Cited by 55 publications

(59 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We objectively quantified the positive rate of Col I by image analysis using ImageJ. In the current study, the Col I‐rich stroma group was defined if >50% of the tumor area showed specifically positive reaction for Col I . The distribution of patients according to the expression level of Col I is shown in Figure .…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Collagen type I induces EGFR‐TKI resistance in EGFR‐mutated cancer cells by mTOR activation through Akt‐independent pathway

Yamazaki

Higuchi

Ishibashi³

et al. 2018

Cancer Science

View full text Add to dashboard Cite

Primary resistance to epidermal growth factor receptor tyrosine kinase inhibitors (EGFR‐TKIs) is a serious problem in lung adenocarcinoma patients harboring EGFR mutations. The aim of this study was to examine whether and how collagen type I (Col I), the most abundantly deposited matrix in tumor stroma, affects EGFR‐TKI sensitivity in EGFR‐mutant cells. We evaluated the EGFR‐TKI sensitivity of EGFR‐mutated cancer cells cultured with Col I. Changes in the activation of downstream signaling molecules of EGFR were analyzed. We also examined the association between the Col I expression in tumor stroma in surgical specimens and EGFR‐TKI response of postoperative recurrence patients with EGFR mutations. Compared to cancer cells without Col I, the survival rate of cancer cells cultured with Col I was significantly higher after EGFR‐TKI treatment. In cancer cells cultured with and without Col I, EGFR‐TKI suppressed the levels of phosphorylated (p‐)EGFR, p‐ERK1/2, and p‐Akt. When compared to cancer cells without Col I, expression of p‐P70S6K, a hallmark of mTOR activation, was dramatically upregulated in cancer cells with Col I. This activation was maintained even after EGFR‐TKI treatment. Simultaneous treatment with EGFR‐TKI and mTOR inhibitor abrogated Col I‐induced resistance to EGFR‐TKI. Patients with Col I‐rich stroma had a significantly shorter progression‐free survival time after EGFR‐TKI therapy (238 days vs 404 days; P < .05). Collagen type I induces mTOR activation through an Akt‐independent pathway, which results in EGFR‐TKI resistance. Combination therapy using EGFR‐TKI and mTOR inhibitor could be a possible strategy to combat this resistance.

show abstract

Section: Methodsmentioning

confidence: 99%

“…In the current study, the Col I-rich stroma group was defined if >50% of the tumor area showed specifically positive reaction for Col I. 30 The distribution of patients according to YAMAZAKI ET AL.…”

Section: Histopathological Analysis and Evaluation Of The Associatimentioning

confidence: 99%

Collagen type I induces EGFR‐TKI resistance in EGFR‐mutated cancer cells by mTOR activation through Akt‐independent pathway

Yamazaki

Higuchi

Ishibashi³

et al. 2018

Cancer Science

View full text Add to dashboard Cite

show abstract

“…TSR assessment is easy, quick and without additional costs. Previous research demonstrated the prognostic value of the TSR in different types of invasive solid tumors, including breast cancer . Most of these studies validated a worse prognosis for patients with stroma‐high tumors.…”

Section: Introductionmentioning

confidence: 96%

The prognostic value of the tumor–stroma ratio is most discriminative in patients with grade III or triple‐negative breast cancer

Vangangelt

Green

Heemskerk

et al. 2020

Intl Journal of Cancer

View full text Add to dashboard Cite

The tumor–stroma ratio (TSR) was evaluated as a promising parameter for breast cancer prognostication in clinically relevant subgroups of patients. The TSR was assessed on hematoxylin and eosin‐stained tissue slides of 1,794 breast cancer patients from the Nottingham City Hospital. An independent second cohort of 737 patients from the Netherlands Cancer Institute to Antoni van Leeuwenhoek was used for evaluation. In the Nottingham Breast Cancer series, the TSR was an independent prognostic parameter for recurrence‐free survival (RFS; HR 1.35, 95% CI 1.10–1.66, p = 0.004). The interaction term was statistically significant for grade and triple‐negative status. Multivariate Cox regression analysis showed a more pronounced effect of the TSR for RFS in grade III tumors (HR 1.89, 95% CI 1.43–2.51, p < 0.001) and triple‐negative tumors (HR 1.86, 95% CI 1.10–3.14, p = 0.020). Comparable hazard ratios and confidence intervals were observed for grade and triple‐negative status in the ONCOPOOL study. The prognostic value of TSR was not modified by age, tumor size, histology, estrogen receptor status, progesterone receptor status, human epidermal growth factor receptor 2 status or lymph node status. In conclusion, patients with a stroma‐high tumor had a worse prognosis compared to patients with a stroma‐low tumor. The prognostic value of the TSR is most discriminative in grade III tumors and triple‐negative tumors. The TSR was not modified by other clinically relevant parameters making it a potential factor to be included for improved risk stratification.

show abstract

“…Therefore, by identifying the stroma present in cancerous tissue, scientists would be able to learn more about the progression of cancer and start early treatments targeting the tumor stroma. Interestingly, recent studies show that the tumorstroma ratio (TSR) derived from image analysis also potentially serves as an independent prognostic factor in many types of cancer [7,9,10]. However, much of such work was done manually.…”

Section: Introductionmentioning

confidence: 99%

“…However, much of such work was done manually. For example, human observers were used for visual estimation of TSR in a previous report [10], which is a tedious and time-consuming task and subject to human bias. Such estimation generally involves tumor image segmentation, which is a critical task in medical image processing.…”

Section: Introductionmentioning

confidence: 99%

A pilot exploratory study of the potentials of deep learning methods in cancer image segmentation and classification

Liu

2019

Preprint

View full text Add to dashboard Cite

Background: Tumor classification and feature quantification from H&E histology images are critical tasks for cancer diagnosis, cancer research, and treatment. However, both tasks involve tedious and time-consuming manual examination of histology images. We explored the possibilities of using deep learning methods to perform segmentation and classification of histology images of cancer tissue for their potential in computeraided tumor diagnosis and other clinical and research applications. Specifically, we tested selected deep learning methods for their performance in the segmentation of stroma and glandular objects in tumor image data and in the classification of tumor images. We automated these tasks to help facilitate downstream tumor image analysis, reduce the labor load of pathologists, and provide them with a second opinion on their analysis. Methods: We modified a patch-based U-Net model and trained it to perform stroma detection and segmentation in cancer tissue. Then the semantic segmentation capabilities of the U-Net model were compared with that of a DeepLabV3+ model. We also explored the possible use of transfer learning to train a patch-based model to classify cancer tissue images as carcinoma and sarcoma and to further classify them as carcinoma subtypes. Results:In spite of the limited dataset available for the pilot study, we found that the unconventional DeepLabV3+ model performed biomedical image segmentation more effectively than U-Net when k-fold cross-validation was utilized, but U-Net still showed promise as an effective and efficient model when we used a customized validation approach. We believe that the DeepLabV3+ model can perform segmentation with even more accuracy if computation resource constraints are removed or if more data is used to augment the result. In terms of tumor classification, our selected models also consistently achieve test accuracies above 80%, with a model trained using transfer learning with VGG-16 network as the feature extractors, or convolutional base performing best. For multi-class tumor subtype classification, we also observed promising test accuracies from our models, and a customized post-processing method provided even higher prediction accuracy on test set images and this method can be further investigated. Conclusions:This pilot exploratory study provided strong evidence for the powerful potentials of deep learning models for segmentation and classification of tumor image data.

show abstract

Tumor-stroma ratio (TSR) in non-small cell lung cancer (NSCLC) patients after lung resection is a prognostic factor for survival

Cited by 55 publications

References 26 publications

Collagen type I induces EGFR‐TKI resistance in EGFR‐mutated cancer cells by mTOR activation through Akt‐independent pathway

Collagen type I induces EGFR‐TKI resistance in EGFR‐mutated cancer cells by mTOR activation through Akt‐independent pathway

The prognostic value of the tumor–stroma ratio is most discriminative in patients with grade III or triple‐negative breast cancer

A pilot exploratory study of the potentials of deep learning methods in cancer image segmentation and classification

Contact Info

Product

Resources

About