Addressing metabolic heterogeneity in clear cell renal cell carcinoma with quantitative Dixon MRI

Yue Zhang, Durga Udayakumar, Ling Cai, Zeping Hu, Payal Kapur, Eun Young Kho, Andrea Pavía-Jiménez, Michael Fulkerson, Alberto Diaz de Leon, Qing Yuan, Ivan E. Dimitrov, Takeshi Yokoo, Jin Ye, Matthew A. Mitsche, Hyeonwoo Kim, Jeffrey G. McDonald, Yin Xi, Ananth J. Madhuranthakam, Durgesh K. Dwivedi, Robert E LenkinskiJeffrey A. Cadeddu, Vitaly Margulis, James Brugarolas, Ralph J. DeBerardinis, Ivan Pedrosa

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

BACKGROUND: Dysregulated lipid and glucose metabolism in clear cell renal cell carcinoma (ccRCC) has been implicated in disease progression, and whole tumor tissue-based assessment of these changes is challenged by the tumor heterogeneity. We studied a noninvasive quantitative MRI method that predicts metabolic alterations in the whole tumor. METHODS: We applied Dixon-based MRI for in vivo quantification of lipid accumulation (fat fraction [FF]) in targeted regions of interest of 45 primary ccRCCs and correlated these MRI measures to mass spectrometry-based lipidomics and metabolomics of anatomically colocalized tissue samples isolated from the same tumor after surgery. RESULTS: In vivo tumor FF showed statistically significant (P < 0.0001) positive correlation with histologic fat content (Spearman correlation coefficient, ρ = 0.79), spectrometric triglycerides (ρ = 0.56) and cholesterol (ρ = 0.47); it showed negative correlation with free fatty acids (ρ = -0.44) and phospholipids (ρ = -0.65). We observed both inter- and intratumoral heterogeneity in lipid accumulation within the same tumor grade, whereas most aggressive tumors (International Society of Urological Pathology [ISUP] grade 4) exhibited reduced lipid accumulation. Cellular metabolites in tumors were altered compared with adjacent renal parenchyma. CONCLUSION: Our results support the use of noninvasive quantitative Dixon-based MRI as a biomarker of reprogrammed lipid metabolism in ccRCC, which may serve as a predictor of tumor aggressiveness before surgical intervention. FUNDING: NIH R01CA154475 (YZ, MF, PK, IP), NIH P50CA196516 (IP, JB, RJD, JAC, PK), Welch Foundation I-1832 (JY), and NIH P01HL020948 (JGM).

Original languageEnglish (US)
JournalJCI insight
Volume2
Issue number15
DOIs
StatePublished - Aug 3 2017

Fingerprint

Renal Cell Carcinoma
Neoplasms
Fats
Lipids
Lipid Metabolism
Metabolomics
Nonesterified Fatty Acids
Disease Progression
Mass Spectrometry
Phospholipids
Triglycerides
Biomarkers
Cholesterol
Pathology
Kidney
Glucose

Keywords

  • Metabolism
  • Oncology

Cite this

@article{d841b19be5544c7d89b0af6198c309c2,
title = "Addressing metabolic heterogeneity in clear cell renal cell carcinoma with quantitative Dixon MRI",
abstract = "BACKGROUND: Dysregulated lipid and glucose metabolism in clear cell renal cell carcinoma (ccRCC) has been implicated in disease progression, and whole tumor tissue-based assessment of these changes is challenged by the tumor heterogeneity. We studied a noninvasive quantitative MRI method that predicts metabolic alterations in the whole tumor. METHODS: We applied Dixon-based MRI for in vivo quantification of lipid accumulation (fat fraction [FF]) in targeted regions of interest of 45 primary ccRCCs and correlated these MRI measures to mass spectrometry-based lipidomics and metabolomics of anatomically colocalized tissue samples isolated from the same tumor after surgery. RESULTS: In vivo tumor FF showed statistically significant (P < 0.0001) positive correlation with histologic fat content (Spearman correlation coefficient, ρ = 0.79), spectrometric triglycerides (ρ = 0.56) and cholesterol (ρ = 0.47); it showed negative correlation with free fatty acids (ρ = -0.44) and phospholipids (ρ = -0.65). We observed both inter- and intratumoral heterogeneity in lipid accumulation within the same tumor grade, whereas most aggressive tumors (International Society of Urological Pathology [ISUP] grade 4) exhibited reduced lipid accumulation. Cellular metabolites in tumors were altered compared with adjacent renal parenchyma. CONCLUSION: Our results support the use of noninvasive quantitative Dixon-based MRI as a biomarker of reprogrammed lipid metabolism in ccRCC, which may serve as a predictor of tumor aggressiveness before surgical intervention. FUNDING: NIH R01CA154475 (YZ, MF, PK, IP), NIH P50CA196516 (IP, JB, RJD, JAC, PK), Welch Foundation I-1832 (JY), and NIH P01HL020948 (JGM).",
keywords = "Metabolism, Oncology",
author = "Yue Zhang and Durga Udayakumar and Ling Cai and Zeping Hu and Payal Kapur and Kho, {Eun Young} and Andrea Pav{\'i}a-Jim{\'e}nez and Michael Fulkerson and {de Leon}, {Alberto Diaz} and Qing Yuan and Dimitrov, {Ivan E.} and Takeshi Yokoo and Jin Ye and Mitsche, {Matthew A.} and Hyeonwoo Kim and McDonald, {Jeffrey G.} and Yin Xi and Madhuranthakam, {Ananth J.} and Dwivedi, {Durgesh K.} and Lenkinski, {Robert E} and Cadeddu, {Jeffrey A.} and Vitaly Margulis and James Brugarolas and DeBerardinis, {Ralph J.} and Ivan Pedrosa",
year = "2017",
month = "8",
day = "3",
doi = "10.1172/jci.insight.94278",
language = "English (US)",
volume = "2",
journal = "JCI insight",
issn = "2379-3708",
publisher = "The American Society for Clinical Investigation",
number = "15",

}

TY - JOUR

T1 - Addressing metabolic heterogeneity in clear cell renal cell carcinoma with quantitative Dixon MRI

AU - Zhang, Yue

AU - Udayakumar, Durga

AU - Cai, Ling

AU - Hu, Zeping

AU - Kapur, Payal

AU - Kho, Eun Young

AU - Pavía-Jiménez, Andrea

AU - Fulkerson, Michael

AU - de Leon, Alberto Diaz

AU - Yuan, Qing

AU - Dimitrov, Ivan E.

AU - Yokoo, Takeshi

AU - Ye, Jin

AU - Mitsche, Matthew A.

AU - Kim, Hyeonwoo

AU - McDonald, Jeffrey G.

AU - Xi, Yin

AU - Madhuranthakam, Ananth J.

AU - Dwivedi, Durgesh K.

AU - Lenkinski, Robert E

AU - Cadeddu, Jeffrey A.

AU - Margulis, Vitaly

AU - Brugarolas, James

AU - DeBerardinis, Ralph J.

AU - Pedrosa, Ivan

PY - 2017/8/3

Y1 - 2017/8/3

N2 - BACKGROUND: Dysregulated lipid and glucose metabolism in clear cell renal cell carcinoma (ccRCC) has been implicated in disease progression, and whole tumor tissue-based assessment of these changes is challenged by the tumor heterogeneity. We studied a noninvasive quantitative MRI method that predicts metabolic alterations in the whole tumor. METHODS: We applied Dixon-based MRI for in vivo quantification of lipid accumulation (fat fraction [FF]) in targeted regions of interest of 45 primary ccRCCs and correlated these MRI measures to mass spectrometry-based lipidomics and metabolomics of anatomically colocalized tissue samples isolated from the same tumor after surgery. RESULTS: In vivo tumor FF showed statistically significant (P < 0.0001) positive correlation with histologic fat content (Spearman correlation coefficient, ρ = 0.79), spectrometric triglycerides (ρ = 0.56) and cholesterol (ρ = 0.47); it showed negative correlation with free fatty acids (ρ = -0.44) and phospholipids (ρ = -0.65). We observed both inter- and intratumoral heterogeneity in lipid accumulation within the same tumor grade, whereas most aggressive tumors (International Society of Urological Pathology [ISUP] grade 4) exhibited reduced lipid accumulation. Cellular metabolites in tumors were altered compared with adjacent renal parenchyma. CONCLUSION: Our results support the use of noninvasive quantitative Dixon-based MRI as a biomarker of reprogrammed lipid metabolism in ccRCC, which may serve as a predictor of tumor aggressiveness before surgical intervention. FUNDING: NIH R01CA154475 (YZ, MF, PK, IP), NIH P50CA196516 (IP, JB, RJD, JAC, PK), Welch Foundation I-1832 (JY), and NIH P01HL020948 (JGM).

AB - BACKGROUND: Dysregulated lipid and glucose metabolism in clear cell renal cell carcinoma (ccRCC) has been implicated in disease progression, and whole tumor tissue-based assessment of these changes is challenged by the tumor heterogeneity. We studied a noninvasive quantitative MRI method that predicts metabolic alterations in the whole tumor. METHODS: We applied Dixon-based MRI for in vivo quantification of lipid accumulation (fat fraction [FF]) in targeted regions of interest of 45 primary ccRCCs and correlated these MRI measures to mass spectrometry-based lipidomics and metabolomics of anatomically colocalized tissue samples isolated from the same tumor after surgery. RESULTS: In vivo tumor FF showed statistically significant (P < 0.0001) positive correlation with histologic fat content (Spearman correlation coefficient, ρ = 0.79), spectrometric triglycerides (ρ = 0.56) and cholesterol (ρ = 0.47); it showed negative correlation with free fatty acids (ρ = -0.44) and phospholipids (ρ = -0.65). We observed both inter- and intratumoral heterogeneity in lipid accumulation within the same tumor grade, whereas most aggressive tumors (International Society of Urological Pathology [ISUP] grade 4) exhibited reduced lipid accumulation. Cellular metabolites in tumors were altered compared with adjacent renal parenchyma. CONCLUSION: Our results support the use of noninvasive quantitative Dixon-based MRI as a biomarker of reprogrammed lipid metabolism in ccRCC, which may serve as a predictor of tumor aggressiveness before surgical intervention. FUNDING: NIH R01CA154475 (YZ, MF, PK, IP), NIH P50CA196516 (IP, JB, RJD, JAC, PK), Welch Foundation I-1832 (JY), and NIH P01HL020948 (JGM).

KW - Metabolism

KW - Oncology

UR - http://www.scopus.com/inward/record.url?scp=85046504207&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046504207&partnerID=8YFLogxK

U2 - 10.1172/jci.insight.94278

DO - 10.1172/jci.insight.94278

M3 - Article

C2 - 28768909

AN - SCOPUS:85046504207

VL - 2

JO - JCI insight

JF - JCI insight

SN - 2379-3708

IS - 15

ER -