Influence of Gene Expression on progression of Renal Cell Carcinoma - - PowerPoint PPT Presentation

Influence of Gene Expression on progression of Renal Cell Carcinoma

Jong Y. Park, PhD, MPH, MS. Associate Member Department of Cancer Epidemiology MoffiH Cancer Center

Disclosure

• None

Objectives

• Descriptive Epidemiology of RCC
• Rationale for outcome biomarkers

predicting prognosis

• Tentative biomarkers for

prognosis of RCC

Trend of incidence and mortality

High Incidence in Men

High Incidence in Men

High Mortality in Men

Distribution of Stage and survival

Risk Factors

12

— Smoking contributes to 25% of RCC cases

• Smoking increased risk 2X

— Environmental:

Cadmium, thorium-di-oxide, petroleum and phenacetin

— Occupational:

Leather tanners, shoe workers, asbestos workers

— Obesity — High blood pressure — Long term dialysis

RCC incidence rates

Genetic Risk Factors

• Von Hippel-Lindau syndrome
• Birt-Hogg-Dube syndrome
• Hereditary papillary RCC
• Hereditary leiomyomatosis

Treatment

• Surgery
• Radiation therapy
• Immunotherapy
• Active surveillance

Biomarkers for prognosis

• ~35-40% of patients with localized

ccRCC at time of surgery will experience metastasis

• Reliable prognostic biomarkers are

urgently needed to improve survival and patient care

Three independent cohorts

• 1. TCGA (n=440)
• 2. Total Cancer Care (TCC) (n=547)
• 3. Moffitt cohort (n=384)

Three independent cohorts

• 1. TCGA (n=440)
• 2. Total Cancer Care (TCC) (n=547)
• 3. Moffitt cohort (n=384)
• 400 candidate genes were selected from

preliminary study and literatures.

• These genes were known to be

associated with RCC progression

TCGA

• A total of 440 ccRCC patients were used in

this work

• 146 out of 400 candidate genes were

associated with survival in TCGA dataset based on the following three criteria

1) Transcripts were identified in any protein-coding region. 2) Transcript sequences have been annotated in GENCODE project. 3) Transcripts were expressed in at least half of the ccRCC samples.

TCC

• Under the Total Cancer Care (TCC)

protocol, 547 ccRCC tumor tissues were collected at the Moffitt Cancer Center.

• Global gene expression assays were

performed using a custom Affymetrix HuRSTA GeneChips.

• 139 out of 400 genes were associated

with recurrence in TCC dataset

Genes associated with progression

TCGA

94

TCC

87 52 169

Genes associated with progression

TCGA

94 (26)

TCC

87 (7) 52 (47) 169 (1)

MoffiH (n=384)

Test 400 candidate genes

Validate 81 idenXfied genes

TCC (n=547) TCGA (n=440)

Study Design

Moffitt cohort

• Historical cohort of 1,368 RCC cases treated

at the MoffiH Cancer Center from 1998 to 2013

• 384 paXents were selected (aggressive vs

indolent)

• Expression level of 81 genes in 384 tumor

Xssues was evaluated with Nanostring assay

81 candidate genes

ABCC1 CDH1 HSPB1 RAPGEF5 ABCG2 CDKN2A IGF1R ROR1 ALOX5 CDKN2B ITGA6 RPL13A AMD1 CEP290 JAG1 SAA1 ANXA2 CORO6 JAK1 SLC22A2 APC CTDSPL JUP SLC5A8 AR CXCR4 KDR SLC6A3 ASPSCR1 CYBA KLF4 SMAD4 AURKA ENO2 MANF SOCS3 AURKB ESR1 MKI67 SOD2 AXL ETS1 MMP3 SPINK1 BCL2 FAM107A MMP9 SPRY1 BHMT FNIP2 MPZL2 TEK BIRC5 FTL MSTO1 TFAP2A BRAF FZD4 MYCN TGFBR3 BTG3 GAPDH MYO6 TGM2 C3 GNL3 PLOD2 THRB CCND1 HDAC1 PPIA TOP2A CCNE1 HLF PRDX6 TPPP CD151 HPX RAD23B UBE2S WASF2

KM1TCGA

<0.05

KM2TCGA

KM3TCGA

KM4TCGA

5 Genes >0.05

KM1TCC

KM2TCC

KM3TCC

KM4TCC

22 Genes >0.05 in either TCC or TCGA

Boxplot1 MoffiH

Boxplot2 MoffiH

Boxplot3 MoffiH

Boxplot4 MoffiH

Boxplot5 MoffiH

30 Genes >0.05 in either TCC or TCGA

Validation data

Validation data

10 candidate biomarkers

ABCG2 AURKA AURKB BCL2 BIRC5 CCNE1 GAPDH ITGA6 SAA1 TOP2A

Summary

— RCC is relatively rare but increasing

incidence

— Associated with tobacco and inherited

disorders

— Surgery is the major curative modality for

Stage I, II, and III

— We identified 10 genes expressed differentially

between indolent and aggressive cases.

42

Acknowledgements

Moffitt Cancer Center Thomas Sellers, PhD Julio Pow-Sang, MD Anders Berglund, PhD Jasreman Dhillon, MD Youngchul Kim, PhD Philip Spiess, MD Wade Sexton, MD Mayo Florida Alex Parker, PhD LSU Hui-Yi Lin, PhD Park’s lab Hyun Park, MS Selina Radlein, MPH Ernest Amankwah, PhD

Natural History

45

— 7% diagnosed incidentally — 45% present with localized disease, 25%

with locally advanced disease, 30% with metastatic disease

— Distant metastases- lung (75%), soft tissue

(36%), bone (20%), liver (18%), skin (8%) and CNS (8%)

— incidence and mortality rates of RCC are

increasing in the United States

Renal Cell Carcinoma

46

— First described by Konig in 1826. — In 1883 Grawitz, noted the fatty content of cancer

cells similar to that of adrenal cells.

— All these tumors arise from Renal tubular epithelium. — accounts for 80–85% of kidney cancer — 2% to 4% increase in incidence per year

Epidemiology

47

— Male predominance (1.6:1.0 M:F) — Highest incidence between age 50-70

• Median age of diagnosis is 66 years
• Median age of death 70 years

— Majority of RCC occurs sporadically — Highest incidence in Scandinavia and North

America, lowest in Africa

Two Approaches for Molecular Epi. Studies

• Candidate Pathway
• Genome Wide Association

Candidate Pathway Approach

Case-Case Study Design

PCa paXents who had a recurrence a_er treatment. PCa patients who are recurrence-free at least 5 years after treatment

Management of Localized disease

51

Localized Disease

Surgery Radio Therapy

MoffiH Cohort

• We select 384 paXents at MoffiH Cancer

Center who were treated surgically for ccRCC between 1990 and 2005, and total RNA was isolated from FFPE Xssue blocks

• The NanoString pladorm was used to quanXfy

gene expression of 81 candidate genes, were idenXfied by TCC and TCGA data.

Results from TCGA and TCC

• Analysis of 400 candidate genes, were selected

from our preliminary study and literatures were evaluated whether they are differenXally expressed and associated with survival using RNA-seq data of 537 ccRCC paXents from the TCGA data portal or Affymatrix data for 547 paXents from TCC. 94 and 87 genes were idenXfied using staXsXcal P-values TCGA and TCC

• data. Among genes idenXfied from two data set,

52 genes were detected from both data.

52 (47)

167 (1)

TCC (n=547) TCGA (n=533) MoffiH

Test 195 candidate genes Validate idenXfied genes Confirm idenXfied genes

MoffiH (n=384)

Test 400 candidate genes Validate 81 idenXfied genes

TCC (n=547) TCGA (n=533)

8 idenXfied genes