Original Article | Published:

An individual coding polymorphism and the haplotype of the SPARC gene predict gastric cancer recurrence

The Pharmacogenomics Journal volume 13, pages 342348 (2013) | Download Citation

Previous Presentation: ASCO 2010 (J Clin Oncol 28:15 s, 2010 (suppl; abstr 4054))

Abstract

The 5-year survival rate for gastric adenocarcinoma (GA) remains only 40% and biomarkers to identify patients at high risk of tumor recurrence are urgently needed. Secreted protein acidic and rich in cysteine (SPARC) is an extracellular matrix glycoprotein that mediates cell matrix interactions, and upregulation of SPARC can promote tumor progression and metastasis. This study investigated whether single-nucleotide polymorphisms (SNPs) in SPARC impact the prognosis of GA. Blood or formalin-fixed, paraffin-embedded tissues were obtained from 137 GA patients at the University of Southern California and Memorial Sloan-Kettering Cancer Center medical facilities. DNA was isolated and five SNPs in the SPARC 3′-untranslated region (UTR) were evaluated by DNA sequencing or PCR-restriction fragment length polymorphism. Associations between SNPs and time to tumor recurrence (TTR) were analyzed using Kaplan–Meier curves, log-rank tests, and likelihood-ratio test within logistic or Cox regression model as appropriate. Patients carrying at least one G allele of the SPARC rs1059829 polymorphism (GG, AG) showed a median TTR of 3.7 years compared with 2.1 years TTR for patients with AA (hazard ratio (HR) 0.57; P=0.033). In a multivariate analysis adjusted for T and N category as covariates and stratified by race, hospital and chemotherapy, patients with at least one SPARC rs1059829 G allele (GG, AG) remained significantly associated with superior TTR than patients with AA genotype (adjusted P=0.026). In addition, patients harboring the G-A-A haplotype had the highest risk of tumor recurrence (HR 1.892; adjusted P=0.016). Our findings suggest that SPARC 3′-UTR SNPs may be useful in predicting GA patients at increased risk of recurrence.

Introduction

Gastric cancer (GC) is currently the fourth most common malignancy worldwide, with 989 600 new cases reported each year. The severity of this disease is reflected in its high mortality rate with 738 000 deaths each year, elevating it to the position of second leading cause of cancer-related deaths worldwide.1 However, although recent advances in diagnosis, multimodal treatment regimens including surgery, radiation, chemo- and antibody-therapy have had some impact, the 5-year overall survival (OS) rate for GC remains low at 40%.2, 3

Although numerous molecular events that lead to GC development have been identified and described, there currently remains a distinct lack of robust and validated predictive and prognostic biomarkers. The identification of such biomarkers will provide clinicians with the power to predict both response to chemotherapy and prognostic markers to evaluate the aggressiveness of the disease and the likelihood of tumor recurrence. The validation and routine clinical implementation of such biomarkers combined with the introduction of novel and efficacious therapies introduce the possibility of a true personalized therapy that involves a simultaneous case-specific analysis of clinical and pathological characteristics, and analysis of a patient's genetic and tumor biomarker profile.

Secreted protein acidic and rich in cysteine (SPARC), also known as osteonectin or BM-40, is an extracellular matrix glycoprotein with high binding affinity to albumin that mediates cell matrix interactions.4 SPARC expression is primarily limited to bone and connective tissues, and tissues undergoing development, remodeling and repair.5 Recent studies have revealed that SPARC functions as a key regulator of critical cellular functions, including proliferation, survival, cell migration and angiogenesis. Recent evidence also indicates that SPARC has a key role in modulating the activities of cytokine and growth factors.6, 7, 8, 9, 10 SPARC upregulation has been reported in a variety of human malignancies, including melanoma,8 breast,11 hepatocellular,12 colorectal,13 lung14 and gastric carcinoma.15, 16, 17 However, the significance of SPARC appears to be variable among tumor types and the tumorigenic effect of SPARC may be dependent on the tumor type and the microenvironment, and these fundamental differences are likely to contribute toward the complex behavior of SPARC in cancer-specific and patient-specific scenarios.16, 18

Upregulation of SPARC expression by cancer cells and reactive stromal cells results in the alteration and redistribution of extracellular matrix components, thereby promoting tumor invasion, progression and metastasis in GC.15, 17, 18 SPARC may also contribute to the invasive and metastatic properties of tumor cells by upregulation of Snail and downregulation of the cell adhesion molecule E-cadherin, thereby promoting the epithelial–mesenchymal transition.8 Recent studies have indicated that in addition to its association with poor prognosis and GC progression, SPARC protein overexpression is associated with poor OS and resistance to cytotoxic chemotherapy, and it has been suggested as a potential biomarker to predict GC progression, metastatic potential and response to chemotherapy.15, 16, 17, 19 The 3′-untranslated region (UTR) of a gene is an important regulatory region that controls mRNA stability and translational efficiency through its interaction with transcription factors,20 ribonucleases and ancillary proteins that can promote or inhibit RNA stability, degradation and cytoplasmic localization. Thus, polymorphisms in the 3′-UTR have the potential to influence the SPARC gene expression levels through a number of alternate mechanisms. Polymorphisms in the 3′-UTR of the SPARC gene have recently been associated with systemic sclerosis21 and osteoporosis.22

Given the important role of SPARC in tumor cell invasion, adhesion and migration, it is highly plausible that functional genetic variations in this gene may direct effects on the progression and prognosis of GC. To test this hypothesis, we selected five common genetic variations in the SPARC gene 3′-UTR and evaluated their individual and combined association with GC recurrence and survival.

Subjects and methods

Study population

The study population included 137 patients with localized (stage Ib–IV) gastric adenocarcinoma (GA). Patients were diagnosed with GA and were recruited between 1992 and 2008 at the University of Southern California/Norris Comprehensive Cancer Center, the Los Angeles County/University of Southern California Medical Center, or the Memorial Sloan-Kettering Cancer Center/Cornell University. All were histopathologically confirmed GA and were treated with surgery alone, or surgery and adjuvant (radio)-chemotherapy. Patient data were collected retrospectively through chart review. The individuals selected for the study provided informed consent for the analysis of molecular correlates, and the study was approved by the Institutional Review Boards of the University of Southern California and Memorial Sloan-Kettering Cancer Center (Table 2).

Single-nucleotide polymorphism selection and genotyping

All common (minor allele frequency 10% in Caucasians) potential functional polymorphisms were searched in the UTR of the SPARC gene, using the HapMap Project database (www.hapmap.org). Potentially functional polymorphisms were identified to meet the following criteria: (a) located in the 3′-UTR and coding regions with amino acid changes, and (b) were shown to be of biological significance according to the literature review.

Either venous blood samples or formalin-fixed, paraffin-embedded normal gastric tissue specimens were obtained. DNA isolation was done by conventional proteinase K-phenol/chloroform–ethanol method using the QIAamp extraction kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s protocol. SPARC genotypes were detected either by direct sequencing. Genotyping was done in a blinded fashion and samples were analyzed in batches with a negative control consisting of a reaction containing no DNA. In case of ambiguous genotype calls due to incomplete restriction enzyme digestion, samples were regenotyped using direct sequencing. Five 3′-UTR polymorphisms were successfully genotyped with call rates >95% and consistent with those expected from the Hardy–Weinberg equilibrium (P>0.05). The genes, reference single-nucleotide polymorphism (SNP) identification numbers, SNP location, function, forward and reverse primer are summarized in Table 1.

Table 1: Analyzed polymorphisms within 3′-UTR of SPARC gene and primer sequences

In silico analysis of SPARC rs1059829 polymorphism

In silico analysis for SPARC rs1059829 polymorphism was done using the functional SNP database (http://compbio.cs.queensu.ca/F-SNP/), which provides integrated information obtained from 16 bioinformatic tools that examine for deleterious effects with respect to each functional category.23, 24

Statistical analysis

The primary endpoints of the current analyses of five SPARC SNPs in patients with localized GA treated with adjuvant therapy included time to tumor recurrence (TTR) and OS. The TTR was calculated as the time from the date of diagnosis of the disease to the date of first observation of tumor recurrence, or until last follow-up if the patient was recurrence-free at that time. The OS was defined as the period from diagnosis to death from any cause, or the last contact if the patient was alive.

The distributions of SPARC SNPs across baseline demographic, clinical and pathological characteristics were assessed using Fisher's exact test. The association between each SPARC SNP with TTR and OS was tested using Kaplan–Meier curves and the log-rank test. Although the mode of inheritance of SPARC SNPs was unknown, we considered the dominant, recessive, co-dominant, or additive genetic model whenever appropriate. Allelic distribution of all polymorphisms in each race/ethnic group was tested for deviation from Hardy–Weinberg equilibrium using a χ2-test with 1° of freedom. Linkage disequilibrium among SPARC was assessed using D′ and r2 values, and the haplotype frequencies were inferred using HaploView version 4.1 (www.broad.mit.edu/mpg/haploview).

The multivariable Cox proportional hazards regression model, including T and N categories as covariates and race, hospital, and type of adjuvant chemotherapy as stratum variables was used to re-evaluate the association between SPARC SNPs, and TTR and OS, considering the imbalances in the distributions of baseline characteristics. An internal validation method, leave-one-out cross validation (LOOCV) was used to evaluate the association between SPARC polymorphisms and TTR in patients with resected GA. The leave-one-out cross validation LOOCV approach provides better estimates of prediction error compared with other resampling methods, especially for small studies.25 The LOOCV has previously been used to select multiple candidate markers for their associations with clinical outcome.26, 27, 28, 29 However, although LOOCV is an informative model, it must be noted that it is not equivalent to external validation with independent data sets. Multiple testing was not adjusted for because of the limited number of polymorphisms and haplotypes investigated in the current study.30 All statistical tests were two-sided at a significance level of 0.05 and conducted using the SAS statistical package version 9.2 (SAS Institute, Cary, NC, USA).

Results

The clinical information and demographic characteristics for the 137 patients with GA analyzed in the study are summarized in Table 2. In a period of up to 3.3 years of follow-up, 61 patients (45%) had tumor recurrence with a probability of 3-year recurrence of 0.52±0.05. A total of 55 of 137 (40%) patients had recurrent disease within the first 3 years after surgery and the median TTR was 2.8 years (95% confidence interval (CI), 2.1–7.0 years). Of the 137 patients, 45 (33%) have died and the median OS of the cohort is 4.7 years (95% CI, 3.8–7.3 years). T stage, N stage and type of chemotherapy were significantly associated with TTR (log-rank P<0.01). No statistically significant association between tested genetic variations and Lauren classification was observed (P>0.05).

Table 2: Baseline demographic and clinical characteristic and clinical outcome in patients with localized GA

Effects of SPARC rs1059829 polymorphisms on clinical outcome—univariate analysis

Genotyping of SPARC rs1059829 was successful in 132 (96%) of the 137 patients. Genotyping was not successful in the remaining five (4%) patients due to a limited quantity and quality of extracted genomic DNA. In the 132 patients successfully genotyped, patients carrying at least one G allele of the SPARC rs1059829 polymorphism (GG, AG) showed a median TTR of 3.7 years compared with 2.1 years TTR for patients with AA (hazard ratio (HR) 0.57; 95% CI, 0.33–0.97; P=0.033, log-rank test). SPARC rs1059829 was not significantly associated with the OS (P=0.5; Table 3).

Table 3: SPARC polymorphisms and time to recurrence and overall survival in patients with localized gastric adenocarcinoma

Multivariate analysis for SPARC rs1059829 and TTR

The Cox proportional hazard model was used to adjust for potential confounding between SPARC rs1059829 polymorphism and TTR. The multivariable model was adjusted for T category and N category as covariates and stratified by race, hospital and type of chemotherapy. GA patients with at least one SPARC rs1059829 G allele (GG, AG) remained significantly associated with superior TTR than patients with AA genotype (adjusted P=0.026; Table 4). Moreover, SPARC G-A-A haplotype remained independently associated with TTR even after adjusting for six significant polymorphisms previously published by our group (PAR-1 2506 ins/del, ES+4349 G>A, IL-8 2251, CD44 rs187115 and rs187116, GRP78 rs391957).31, 32, 33

Table 4: A multivariate analysis of SPARC polymorphisms and time to recurrence and overall survival in patients with resected gastric cancer

Haplotype analysis

Three tested SPARC polymorphisms (rs1053411, rs1059279, rs1059829) were in linkage disequilibrium in our study population (D′0.96 and r20.31). The most common three haplotypes (those with a 5% frequency) comprise 98.1% of the total predicted haplotype variation. TTR differed significantly according the patients’ SPARC haplotypes (Table 4). Patients harboring the G-A-A haplotype were at highest risk to develop tumor recurrence (HR: 1.985; 95% CI, 1.167–3.377) compared with patients with the most prevalent G-A-G haplotype (adjusted P=0.011). Moreover, we subsequently tested whether our recently published data concerning CD4433 is independent of the SPARC associations described in the current manuscript. We included both SPARC and CD44 haplotypes in the multivariate model and found that both remained significant (data not shown). In addition, patients with G-A-A haplotype were at highest risk of death (HR 1.672; 95% CI, 0.941, 2.972), however, did not reach statistical significance (adjusted P=0.08; Figures 1a and b). The other tested polymorphisms did not show any linkage disequilibrium.

Figure 1
Figure 1

Time to tumor recurrence (a) and overall survival (b) by G-A-G, C-C-A and G-A-A haplotypes of SPARC rs1054204, rs1059279, and rs1059829 polymorphisms in localized gastric adenocarcinoma patients. All censored patients and those who showed tumor recurrence are accounted for.

Leave-one-out cross validation

LOOCV was performed using a single observation from the original sample as the validation data, and the remaining observations as the training data (124 of 136 leave-one-out data sets). SPARC rs1059829 polymorphism in the dominant model was significantly associated with TTR in both univariate and multivariate analysis.

Analysis of other tested SPARC polymorphisms

None of the other tested polymorphisms showed a statistically significant relationship between TTR and OS (Tables 3 and 4).

Discussion

In this translational GC study, we investigated the effects of five potentially functional SPARC polymorphisms on clinical outcome of patients with GA. The results provide evidence that SPARC rs1059829 polymorphism and G-A-A haplotype of 3′-UTR SNPs may serve as potential prognostic determinants for GA recurrence. Importantly, these results remained significant after adjusting for other potential predictors of patient outcome, including CD44 haplotypes previously evaluated in this patient cohort. This represents the first report showing that the individual coding polymorphism rs1059829 and SPARC gene haplotype may serve as prognostic marker for localized GA patients who were treated with surgery alone or surgery and adjuvant (radio)-chemotherapy.

A series of epidemiological studies have shown that high SPARC protein expression has a key role in GC progression, and evidence of the influence on prognosis has accumulated recently.15, 17 Maeng et al.34 recently reported that SPARC is highly expressed in reactive stroma associated with invasive differentiated adenocarcinomas, and that it may serve as a potential prognostic marker of GC progression and advanced stage disease. The differential expression of SPARC in the epithelial and stromal compartment described by Wewer et al.35 further suggests that both cell types have the capacity to produce SPARC. However, stromal cells most likely proliferate in response to the presence of the tumor cells.

Recent evidence has demonstrated clear roles for SPARC in the regulation of key molecules involved in cell adhesion and epithelial to mesenchymal transition. Specifically, the regulation of Snail and E-cadherin levels by SPARC during oncogenic transformation of melanocytes provided important insights into the pathways by which SPARC exerts it's tumorigenic properties and drives aggressive tumor progression.8 In addition, SPARC also interacts with matrix metalloproteinase expression and cytoskeleton architecture, thereby facilitating the invasion and metastasis of tumor cells.36, 37, 38

In the present study, we found a significant association between patients carrying at least one G allele (A/G, G/G) of the SPARC rs1059829 polymorphism in the 3′-UTR with a significantly lower risk of tumor recurrence than patients with homozygous A allele (HR 0.447; 95% CI, 0.22–0.91; adjusted P=0.026). Furthermore, we constructed a genetic haplotype across the 3′-UTR of SPARC showing a 1.99-fold (95% CI, 1.17–3.38) increased risk for tumor recurrence for patients carrying G-A-A haplotype after adjusting for other potential predictors of the patients’ outcome (adjusted P=0.011). Finally, the SPARC rs1059829 polymorphism and G-A-A haplotype demonstrated a clear trend towards a statistically significant increased risk of death (HR 1.672; 95% CI, 0.94–2.97; adjusted P=0.08).

Currently, the precise functional and biological significance by which the SPARC rs1059829 exerts its influence on tumor recurrence has not yet been elucidated and this remains a significant limitation when interpreting the results of the current study. However, in silico analysis showed that the A allele of SPARC rs1059829 polymorphism creates a CCAAT enhancer-binding protein (C/EBPbeta) binding site, which is not present in the minor allele (G allele). The C/EBPbeta is an important regulator of cell growth, differentiation and in promoting tumor invasiveness. The C/EBPbeta is a member of the transcriptional factor family consisting of six functionally and structurally related basic leucine zipper DNA-binding proteins. The C/EBPbeta gene is located on the long arm of chromosome 20 (20q), where DNA copy number amplification has been observed in a wide variety of cancers, including GC.39 Indeed, the most frequent amplified chromosomal region in GCs was observed at 20q.40 Although replications in independent cohorts and further functional evaluations of SPARC polymorphisms are still needed, in silico analysis supports our preliminary findings that SPARC rs1059829 polymorphism is associated with GC recurrence.

Recent translational studies have proposed SPARC as a promising novel target for the treatment of GC. Although high SPARC protein expression is generally associated with poor prognosis in GA, SPARC also serves as the target for albumin-based chemotherapies, such as nanoparticle albumin-bound paclitaxel (nab-paclitaxel), because of its albumin-binding activity. Desai et al.41 showed that head and neck cancer patients with high SPARC protein expression had significantly higher response to nab-paclitaxel than SPARC-negative patients (83% versus 25%). Moreover, Inoue et al.42 suggested SPARC as a potentially useful target candidate for cancer immunotherapy in various cancers, including GA. Jeung et al.19 suggested in a randomized phase II study high SPARC protein expression as potential predictive marker of resistance to docetaxel in primary gastric tumors. This recent data and the results of the present study suggest that SPARC may be both a prognostic and predictive marker, with the power to identify patients with a high risk of recurrence and the ability to identify those patients that may or may not respond favorably to therapy.

In summary, our preliminary study provides the first evidence that polymorphisms in the 3′-UTR of SPARC have an important role in tumor progression of localized GA patients treated with surgery alone, or surgery and adjuvant (radio)-chemotherapy. In addition, these data may help to predict response in a subgroup of patients treated with nab-paclitaxel. Further validation of our hypothesis-generating findings in prospective biomarker-embedded clinical trials is needed.

References

  1. 1.

    , , , , , . Global cancer statistics. CA Cancer J Clin 2011; 61: 212–236.

  2. 2.

    , , , , , et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med 2006; 355: 11–20.

  3. 3.

    , , , , , et al. Extended lymph node dissection for gastric cancer: who may benefit? Final results of the randomized Dutch gastric cancer group trial. J Clin Oncol 2004; 22: 2069–2077.

  4. 4.

    , , . Characterization of a novel serum albumin-binding glycoprotein secreted by endothelial cells in culture. J Biol Chem 1984; 259: 3993–4007.

  5. 5.

    , , , , . Distribution of SPARC in normal and neoplastic human tissue. J Histochem Cytochem 1995; 43: 791–800.

  6. 6.

    , . SPARC, a matricellular protein that functions in cellular differentiation and tissue response to injury. J Clin Invest 2001; 107: 1049–1054.

  7. 7.

    , , , , . The extracellular glycoprotein SPARC interacts with platelet-derived growth factor (PDGF)-AB and -BB and inhibits the binding of PDGF to its receptors. Proc Natl Acad Sci USA 1992; 89: 1281–1285.

  8. 8.

    , , , , , et al. SPARC represses E-cadherin and induces mesenchymal transition during melanoma development. Cancer Res 2006; 66: 7516–7523.

  9. 9.

    , , , , , et al. Secreted protein acidic, rich in cysteine (SPARC), mediates cellular survival of gliomas through AKT activation. J Biol Chem 2004; 279: 52200–52209.

  10. 10.

    , , , , , et al. SPARC-induced increase in glioma matrix and decrease in vascularity are associated with reduced VEGF expression and secretion. Int J Cancer 2008; 122: 2735–2743.

  11. 11.

    , , , , . Transcriptional upregulation of SPARC, in response to c-Jun overexpression, contributes to increased motility and invasion of MCF7 breast cancer cells. Oncogene 2002; 21: 7077–7091.

  12. 12.

    , , , , , et al. Overexpression of SPARC obliterates the in vivo tumorigenicity of human hepatocellular carcinoma cells. Int J Cancer 2010; 126: 2726–2740.

  13. 13.

    , . A novel interaction between procaspase 8 and SPARC enhances apoptosis and potentiates chemotherapy sensitivity in colorectal cancers. J Biol Chem 2007; 282: 34457–34467.

  14. 14.

    , , , , , et al. Enhanced expression of SPARC/osteonectin in the tumor-associated stroma of non-small cell lung cancer is correlated with markers of hypoxia/acidity and with poor prognosis of patients. Cancer Res 2003; 63: 5376–5380.

  15. 15.

    , , , , . Overexpression of SPARC gene in human gastric carcinoma and its clinic-pathologic significance. Br J Cancer 2004; 91: 1924–1930.

  16. 16.

    , , , , , et al. Downregulation of SPARC expression decreases gastric cancer cellular invasion and survival. J Exp Clin Cancer Res 2010; 29: 59.

  17. 17.

    , , , , . SPARC is associated with gastric cancer progression and poor survival of patients. Clin Cancer Res 2010; 16: 260–268.

  18. 18.

    , , , , , . The role of the matricellular protein SPARC in the dynamic interaction between the tumor and the host. Cancer Metastasis Rev 2008; 27: 691–705.

  19. 19.

    , , , , , et al. A randomized phase 2 study of docetaxel and S-1 versus docetaxel and cisplatin in advanced gastric cancer with an evaluation of SPARC expression for personalized therapy. Cancer 2011; 117: 2050–2057.

  20. 20.

    , . When the message goes awry: disease-producing mutations that influence mRNA content and performance. Cell 2001; 107: 411–414.

  21. 21.

    , , , , , et al. Association of novel polymorphisms with the expression of SPARC in normal fibroblasts and with susceptibility to scleroderma. Arthritis Rheum 2002; 46: 2990–2999.

  22. 22.

    , , , , . Osteonectin/SPARC polymorphisms in Caucasian men with idiopathic osteoporosis. Osteoporos Int 2008; 19: 969–978.

  23. 23.

    , . F-SNP: computationally predicted functional SNPs for disease association studies. Nucleic Acids Res 2008; 36(Database issue): D820–D824.

  24. 24.

    , . An integrative scoring system for ranking SNPs by their potential deleterious effects. Bioinformatics 2009; 25: 1048–1055.

  25. 25.

    , , . Prediction error estimation: a comparison of resampling methods. Bioinformatics 2005; 21: 3301–3307.

  26. 26.

    , , , , , et al. Phosphorylated ERalpha, HIF-1alpha, and MAPK signaling as predictors of primary endocrine treatment response and resistance in patients with breast cancer. J Clin Oncol 2009; 27: 227–234.

  27. 27.

    , , , , , et al. Effectiveness of gene expression profiling for response prediction of rectal adenocarcinomas to preoperative chemoradiotherapy. J Clin Oncol 2005; 23: 1826–1838.

  28. 28.

    , , , , , et al. EGF61 polymorphism predicts complete pathologic response to cetuximab-based chemoradiation independent of KRAS status in locally advanced rectal cancer patients. Clin Cancer Res 2011; 17: 5161–5169.

  29. 29.

    , , , , , et al. Postoperative nomogram predicting the 10-year probability of prostate cancer recurrence after radical prostatectomy. J Clin Oncol 2005; 23: 7005–7012.

  30. 30.

    . No adjustments are needed for multiple comparisons. Epidemiology 1990; 1: 43–46.

  31. 31.

    , , , , , et al. Genetic variations in angiogenesis pathway genes associated with clinical outcome in localized gastric adenocarcinoma. Ann Oncol 2010; 21: 78–86.

  32. 32.

    , , , , , et al. GRP78 promoter polymorphism rs391957 as potential predictor for clinical outcome in gastric and colorectal cancer patients. Ann Oncol 2011; 22: 2431–2439.

  33. 33.

    , , , , , et al. Germline polymorphisms in genes involved in the CD44 signaling pathway are associated with clinical outcome in localized gastric adenocarcinoma. Int J Cancer 2011; 129: 1096–1104.

  34. 34.

    , , , , , et al. Osteonectin-expressing cells in human stomach cancer and their possible clinical significance. Cancer Lett 2002; 184: 117–121.

  35. 35.

    , , , , . Osteonectin/SPARC/BM-40 in human decidua and carcinoma, tissues characterized by de novo formation of basement membrane. Am J Pathol 1988; 132: 345–355.

  36. 36.

    , , , , , . SPARC/osteonectin induces matrix metalloproteinase 2 activation in human breast cancer cell lines. Cancer Res 1998; 58: 5529–5536.

  37. 37.

    , , , , . Differential expression of osteonectin/SPARC during human prostate cancer progression. Clin Cancer Res 2000; 6: 1140–1149.

  38. 38.

    , , , . SPARC, a secreted protein associated with morphogenesis and tissue remodeling, induces expression of metalloproteinases in fibroblasts through a novel extracellular matrix-dependent pathway. J Cell Biol 1993; 121: 1433–1444.

  39. 39.

    , , , . Overexpression of CEBPbeta correlates with decreased TFF1 in gastric cancer. Oncogene 2006; 25: 643–649.

  40. 40.

    , , , , , et al. Genetic differences between adenocarcinomas arising in Barrett′s esophagus and gastric mucosa. Gastroenterology 2001; 121: 592–598.

  41. 41.

    , , , . SPARC Expression Correlates with Tumor Response to Albumin-Bound Paclitaxel in Head and Neck Cancer Patients. Transl Oncol 2009; 2: 59–64.

  42. 42.

    , , , , , et al. Identification of SPARC as a candidate target antigen for immunotherapy of various cancers. Int J Cancer 2010; 127: 1393–1403.

Download references

Acknowledgements

This work was funded by the NIH Grant 5 P30CA14089-27I and the Dhont Family Foundation. It was performed in the Sharon A. Carpenter Laboratory at USC/Norris Comprehensive Cancer Center and in memory of David Donaldson.

Author information

Author notes

    • T Winder

    Supported in part by a Research Grant of the Austrian Society of Hematology and Oncology and the ‘Kurt und Senta-Herrmann foundation’.

Affiliations

  1. Division of Medical Oncology, University of Southern California/Norris Comprehensive Cancer Center, Keck School of Medicine, Los Angeles, CA, USA

    • T Winder
    • , W Zhang
    • , Y Ning
    • , P Bohanes
    • , A Gerger
    •  & H-J Lenz
  2. Department of Pathology, University of Southern California/Norris Comprehensive Cancer Center, Keck School of Medicine, Los Angeles, CA, USA

    • P M Wilson
  3. Department of Preventive Medicine, University of Southern California/Norris Comprehensive Cancer Center, Keck School of Medicine, Los Angeles, CA, USA

    • D Yang
    •  & H-J Lenz
  4. Gastrointestinal Oncology Service, Memorial Sloan-Kettering Cancer Center, Cornell University, New York, NY, USA

    • D G Power
    •  & M Shah
  5. Department of Pathology, Memorial Sloan-Kettering Cancer Center, Cornell University, New York, NY, USA

    • L H Tang

Authors

  1. Search for T Winder in:

  2. Search for P M Wilson in:

  3. Search for D Yang in:

  4. Search for W Zhang in:

  5. Search for Y Ning in:

  6. Search for D G Power in:

  7. Search for P Bohanes in:

  8. Search for A Gerger in:

  9. Search for L H Tang in:

  10. Search for M Shah in:

  11. Search for H-J Lenz in:

Competing interests

The authors declare no conflict of interest.

Corresponding author

Correspondence to H-J Lenz.

About this article

Publication history

Received

Revised

Accepted

Published

DOI

https://doi.org/10.1038/tpj.2012.11

Further reading