Offer, Steven M.

Abstract Abstract 3678 The JAK/STAT pathway is becoming a relevant target for lymphoma therapy and inhibitors of the pathway are now in clinical trial. It would be ideal to identify patients for trial by assessment of pathway activation. We analyzed the primary tumors from 50 patients with diffuse large B cell lymphoma (DLBCL) using antibody against phosphorylated tyrosine signal transducer and activator of transcription 3 (STAT3) revealed that 52% (21/40) of DLBCL tumors express tyrosine phosphorylated STAT3 (pSTAT3). There are several potential mechanisms of activation of the JAK/STAT pathway. Protein tyrosine phosphatases such as SHP1 are key negative regulators of this pathway. SHP-1 is also considered a tumor suppressor gene because of its ability to antagonize the action of tyrosine kinases. Persistent activation of STAT3 seems to play a critical role in the pathogenesis of several malignancies. In an attempt to determine mechanism of STAT3 phosphorylation, we analyzed SHP1 expression and its correlation to STAT3 phosphorylation in DLBCL tumors. We analyzed SHP1 expression by immunohistochemistry (IHC) in a tissue microarray composed of 27 DLBCL tumor samples. SHP1 was found silenced in 33% (7/27) of DLBCL cases. To confirm our IHC data SHP1 mRNA expression was also analyzed in DLBCL patients (n=7) and CD19+ normal B cells (n=3) using reverse transcriptase polymerase chain reaction (RT-PCR). SHP1 expression was found silenced in 57% (4/7) of cases, however normal B cells expressed a significant amount of SHP1 mRNA transcripts. We next sought to determine the mechanism of SHP1 silencing in these patients. DNA methylation is the most common mechanism of gene silencing and gene inactivation in human cancers and methylation of a cluster of CPG sites near the proximal SHP1 promoter 2 has been shown in myeloma and T cell lymphoma cells. We analyzed the frequency of SHP1 DNA methylation in 40 DLBCL patients and in DLBCL cell lines by analyzing the proximal promoter regions for SHP1 with methylation specific (MSP) and unmethylation specific (UMSP) PCR primers. We found SHP1 methylation in 2.5% (1/40) DLBCL tumors and none of the DLBCL cell lines tested (OCILy3, OCILy10, SUDHL2, SUDHL6, OCILy19 and OCILy1) were positive for methylation. Furthermore, bisulphite sequencing of the MSP product confirmed methylated cytosines in the one MSP-positive DLBCL patient sample and in the positive control (methylated DNA). We next confirmed our MSP methylation results in the 40 DLBCL tumors DNA with a high resolution melting (HRM) methylation array for accurate detection of changes in the CpG methylation. The primers were designed to flank CpG islands, not to specifically amplify (converted) methylated or unmethylated DNA (as in MSP PCR). Data were normalized to an unmethylated control. Based on the melting profile, there was no methylation in the controls but detectable methylation (>5%) in the SHP1 promoter region was found in 10% (4/40) of DLBCL samples. The one sample that was positive with MSP PCR was also >50% methylated by HRM. Next we analyzed whether there is any correlation between SHP1 expression and tyrosine phoshorylated STAT3 (pSTAT3). In the 27 cases of DLBCL studied, 16 were pSTAT3 negative and 11 were pSTAT3 positive by IHC. 50% (8/16) of the pSTAT3 negative cases had very high expression of SHP1. Similarly, 45 % (5/11) of pSTAT3 positive cases were SHP1 negative consistent with loss of SHP1 gene expression. These data suggest that loss of SHP1 expression is one of several potential mechanisms that can contribute to increased tyrosine STAT3 activation in DLBCL. In addition, these data suggest that CPG island promoter hypermethylation in the SHP1 promoter is not associated with SHP1 gene silencing in DLBCL tumors. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 2655 Diffuse large B cell lymphoma (DLBCL) has been classified into two distinct molecular subtypes: germinal center B cell–like (GCB), non-germinal centre-like (non-GCB). To improve outcomes for DLBCL patients, it is necessary to identify additional novel targets within GCB and non-GCB classifications to further stratify patients for prognosis and assist in choosing therapy for the individual patient. We have recently demonstrated that STAT3 activation is frequent in the DLBCL tumors, however the exact molecular mechanism(s) of STAT3 activation in DLBCL tumors are not known. Molecular mechanisms such as epigenetic silencing of negative regulators of the STAT3 pathway such as protein tyrosine phosphatase 6 (PTPN6) may contribute to STAT3 activation. We aimed to learn whether PTPN6 was expressed in GCB and non-GCB DLBCL, and if so, how that expression correlated with STAT3 activation and prognosis. We first performed epigenetic studies of PTPN6 in 38 DLBCL tumors and 6 DLBCL cell lines by methylation specific (MSP) PCR and high resolution melting array (HRM) methods. Surprisingly, PTPN6 promoter hypermethylation (0/38) was not detected in patient sample and cell lines by both the methods. Since the MSP PCR technique yields qualitative rather than quantitative data, we next performed pyrosequencing on the 38 DLBCL samples, and found results consistent with the MSP PCR analysis. Our data conclusively demonstrate that PTPN6 hypermethylation is absent in DLBCL tumors. We next sequenced the 600 bp upstream of the transcription initiation site of PTPN6 promoter 2 and 1 in 10 DLBCL tumors. We did not detect any point mutations in the promoter 2 and 1 core regions. Since PTPN6 promoter hypermethylation and mutations were absent in DLBCL tumors, we determined the expression level of the PTPN6 protein in 40 DLBCL tumors by molecular subtype. Formalin fixed paraffin-embedded DLBCL tumor samples were stained by immunohistochemistry (IHC) for the determination of molecular subtype using the Hans algorithm and the detection of PTPN6 expression. Using a threshold of ≥30%, 75% (30/40) of cases were PTPN6 positive with various levels of expression: 11 cases had 30–80% of tumor cells staining positive and 19 had >80% of cells PTPN6 positive. PTPN6 expression by IHC was only correlated with higher IPI scores and a trend towards a shorter event free survival (EFS) (p=0.07). Within the 29 GCB tumors 69% (20/29) were PTPN6 positive; 100% (10/10) of non-GCB cases were PTPN6 positive. These data clearly suggest that PTPN6 expression is found in both GCB and non-GCB with the latter being uniformly positive (p=0.03) and PTPN6 negative cases being uniformly GCB. PTPN6 mRNA and protein was detected in all three ABC lines (LY3, Ly10, DHL2). Within the GCB lines (DHL6, Ly1 and Ly19) DHL6 was weakly positive and Ly1 and Ly19 were negative for PTPN6 mRNA and protein. Furthermore, within the GCB group PTPN6 positive cases had inferior EFS as compared to PTPN6 negative cases. In the non-GCB group all cases were PTPN6 positive with an EFS similar to PTPN6 positive GCB cases. The role of PTPN6 in the persistent activation of the STAT3 pathway in DLBCL patients has not been investigated. We hypothesized that tumors with activated STAT3 would have loss of PTPN6. Interestingly, this hypothesis was disproven. Within the 15-pSTAT3 positive cases 12 (80%) were PTPN6 positive. Conversely, 26% (6/23) of the pSTAT3 negative cases were PTPN6 negative. The distribution of pSTAT3 and PTPN6 by IHC in samples was evaluated in both GCB and non-GCB groups. Within the PTPN6+/pSTAT3+ group out of the 12 cases most were non-GCB (8/12; 66%). However, within the PTPN6-/pSTAT3- group all the 5 cases were GCB (5/5; 100%). Survival analysis revealed that the groups with the best EFS were those with PTPN6-/pSTAT3- tumors (n=5); those with PTPN6+/pSTAT3+ group (n=12) had the shortest EFS; and those with PTPN6+/pSTAT3- tumors (n=17) being intermediate between the other groups. In summary, we have demonstrated for the first time that PTPN6 is highly expressed in DLBCL tumors, and a common abnormality in non-GCB subtypes, which is positively correlated with activated STAT3. PTPN6 expression in the DLBCLs is not regulated through SHP1 promoter hypermethylation or point mutations. The finding that SHP1 loss was found only in GCB cases and was especially favorable should be explored further and may provide an important new stratification factor for future DLBCL studies. Disclosures: No relevant conflicts of interest to declare.

Abstract Colorectal cancer (CRC) is the third most common cancer and is expected to cause approximately 50,000 deaths in the US in 2017. The antimetabolite 5-fluorouracil (5-FU) is the cornerstone of first-line adjuvant chemotherapy regimens for CRC. Resistance to 5-FU-based chemotherapies develops over time and is the greatest obstacle to effective management in this disease. However, the resistance mechanisms of 5-FU are largely unknown. Approximately 50% of the CRC cases have activating mutations in KRAS or BRAF. Based on recent clinical data, it is now appreciated that KRAS/BRAF mutations are associated with worse survival in 5-FU-based chemotherapies for CRC. Therefore, there is an urgent unmet need for improved therapeutic options to treat mutant KRAS/BRAF tumors. We analyzed the differential activation status of enhancers in KRAS/BRAF mutant and wildtype CRC cell lines. Enhancers that associated with genes regulating pyrimidine metabolism were significantly enriched in KRAS/BRAF mutant CRC cells. Among them, the differential activation of an enhancer located ~400kb upstream of the DPYD promoter was most significant. By using CRISPR-dCas9 system, specifically targeting of p300 to the enhancer significantly increased DPYD expression, and Ezh2 significantly decreased expression. Knockout (KO) of this enhancer resulted in reductions in DPD expression and increased resistance to 5-FU. In mouse xenograft studies, enhancer-KO cells responded to doses of 5-FU that were otherwise ineffective against control cells. Treating KRAS mutant CRC cells with CBP30, H3K27 acetyl-transferase inhibitor, significantly reduced DPYD expression. Synergistic interactions was identified between CBP30 and 5-FU. Finally, mouse xenograft studies strongly suggest that inhibition of H3K27ac may significantly improve 5-FU effectiveness against KRAS mutant tumors. In summary, the present studies advance our understanding of how mutant KRAS/BRAF epigenetically regulates DPYD expression and results in 5-FU resistance in CRC. We showed that clinical treatment of CRC might be significantly improved by the development of future pharmacologic approaches targeting enhancers that are activated by mutant KRAS/BRAF. Citation Format: Rentian Wu, Qian Nie, Phillip A. Kubica, Zhiquan Wang, Shikshya Shrestha, Colbren S. Trogstad-Isaacson, Calvin R. Jerde, Erin E. Tapper, Kelly J. Bouchonville, Steven M. Offer, Robert B. Diasio. Targeting the mutant KRAS/BRAF-induced enhancer as an approach to overcome resistance to 5-fluorouracil-based chemotherapy in colorectal cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 911.

Abstract 5-Fluorouracil (5-FU) is a cornerstone therapy for the treatment of several aggressive tumors. However, nearly 1/3 of patients receiving 5-FU experience severe (grade 3+) adverse 5-FU toxicity. Genetic variations in dihydropyrimidine dehydrogenase (DPD; DPYD gene) are major contributors of 5-FU toxicity. With continued progress in next-generation sequencing technologies, the number of reported genetic variants is substantially increasing; to date, there are more than 450 reported missense variants in the DPYD gene. Only a handful of those variants, including *2A, I560S, D949V, and rs75017182, have been established as predictors of toxicity. Because most of the remaining variants are unstudied, mainly due to their rare occurrence and unequal racial distribution, it is challenging to conduct statistically powered clinical association studies. Therefore, in vitro and computational methods to identify potential toxicity-associated variants have major appeal. The primary goal of the current study was to develop a highly accurate gene-specific in silico classifier to predict deleterious DPYD variants. For this purpose, using a recombinant cellular model, we characterized 69 missense DPYD variants reported in online databases. We then utilized DPD activity data from the 69 variants in combination with data from 87 previously characterized variants (see Offer, SM et al. Cancer Research 2013 & 2014) to develop the DPYD variant classifier model (DPYD-Varifier). DPD protein-specific biochemical and structural features and amino acid conservation scores were integrated into the model's random forest classifier-based algorithm. DPYD-Varifier showed improved accuracy (85% accurate) compared to other in silico tools. Additionally, the model provided mechanistic insights into alterations of DPD function conferred by DPYD variants. In general, deleterious variants were located in closer proximity to important binding/coordinating sites of the DPD protein. Additionally, DPYD-Varifier allowed us to estimate the cumulative allele frequency of all reported deleterious variants in various geographical/racial groups, which can substantially add to our knowledge of DPD deficiency and 5-FU toxicity observed in these populations. In conclusion, the integration of in vitro and accurate computational data from our study permitted classification of relevant DPYD variants, which could be useful to more effectively individualize 5-FU therapy in the future. Citation Format: Shikshya Shrestha, Cheng Zhang, Calvin R. Jerde, Hu Li, Steven M. Offer, Robert B. Diasio. DPYD-Varifier, a computational model to identify 5-FU toxicity-associated DPYD variants [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3894.

Abstract 5-Fluorouracil (5-FU) has remained a widely prescribed anti-cancer drug for decades. However, 5-FU–based treatments show inter-individual variability, with up to 34% of 5-FU–treated patients developing severe (grade 3+) adverse toxicity. Dihydropyrimidine dehydrogenase (DPD; DPYD gene) is the rate-limiting enzyme of 5-FU catabolism and converts 80–85% of 5-FU into its inactive metabolites. Three deleterious genetic variations in DPYD are well-established predictors of severe 5-FU toxicity (i.e., *2A, p. I560S, and p. D949V); however, these variants only explain 30–35% of adverse cases. Recently, an intronic variant (rs75017182) has been suggested to contribute to 5-FU–related toxicity by promoting the alternative splicing of DPYD. However, clinical studies are conflicting as to the degree to which rs75017182 affects DPD enzyme function, and the true contribution of the rs75017182 variant towards DPD function and 5-FU toxicity remains unclear and inconclusive. In the present study, we directly examined the effect of the intronic SNP rs75017182 on DPYD mRNA splicing, DPYD expression, and DPD enzyme function. We genotyped 3950 healthy Caucasian volunteers from the Mayo Clinic Biobank to identify the rs75017182, *2A, p.I560S, and p.D949V variants. From the initial cohort, 204 volunteers who were rs75017182 heterozygous carriers were recruited as our study population. Alternative splicing and DPD activity in rs75017182 carriers were measured. The findings were confirmed using a novel mini-gene reporter system in vitro. A moderate, but significant, reduction (30%) was detected in canonically spliced DPYD expression in rs75017182 carriers compared to non-carriers. A correlative reduction (35%) in DPD enzyme activity was observed in those carriers, which was similar to that of D949V carriers (31%). The results demonstrated an association of this deep intronic variant with decreased DPD activity and suggested that rs75017182 may be a predictor of 5-FU toxicity similar to D949V. Citation Format: Qian Nie, Shikshya Shrestha, Erin E. Tapper, Colbren S. Trogstad-Isaacson, Kelly J. Bouchonville, Steven M. Offer, Robert B. Diasio. Quantitative contribution of rs75017182 to dihydropyrimidine dehydrogenase mRNA splicing and enzyme activity [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3893.

Deleterious variants in dihydropyrimidine dehydrogenase (DPD, DPYD gene) can be highly predictive of clinical toxicity to the widely prescribed chemotherapeutic 5‐fluorouracil (5‐FU). However, there are very limited data pertaining to the functional consequences of the >450 reported no‐synonymous DPYD variants. We developed a DPYD‐specific variant classifier (DPYD‐Varifier) using machine learning and in vitro functional data for 156 missense DPYD variants. The developed model showed 85% accuracy and outperformed other in silico prediction tools. An examination of feature importance within the model provided additional insight into functional aspects of the DPD protein relevant to 5‐FU toxicity. In the absence of clinical data for unstudied variants, prediction tools like DPYD‐Varifier have great potential to individualize medicine and improve the clinical decision‐making process.

The purpose of this guideline is to provide information for the interpretation of clinical dihydropyrimidine dehydrogenase ( DPYD ) genotype tests so that the results can be used to guide dosing of fluoropyrimidines (5‐fluorouracil and capecitabine). Detailed guidelines for the use of fluoropyrimidines, their clinical pharmacology, as well as analyses of cost‐effectiveness are beyond the scope of this document. The Clinical Pharmacogenetics Implementation Consortium (CPIC ® ) guidelines consider the situation of patients for which genotype data are already available (updates available at https://cpicpgx.org/guidelines/guideline‐for‐fluoropyrimidines‐and‐dpyd/ ).

Abstract The antimetabolite 5-fluorouracil (5-FU) is one of the most widely used chemotherapy drugs. Dihydropyrimidine dehydrogenase (DPD) initiates the catabolic degradation of 5-FU to inactive metabolites and is widely accepted as a major determinant of 5-FU response and toxicity. While genetic variants in DPYD have been shown to variants may affect 5-FU metabolism, they are rare events and do not completely explain the reported variability in DPD function or the resultant differences in treatment response. Here, we report that DPYD expression is epigenetically regulated by histone modification at relevant promoter and enhancer regions. Using both chemical inhibitors and genetic approaches, we show that that H3K27 tri-methylation (H3K27me3) at the DPYD promoter is regulated by the epigenetic modifiers Ezh2 and UTX. Promoter H3K27me3 suppresses DPYD expression by inhibiting the binding of the transcription factor PU.1 bindingto the promoter and ,increases sensitivity leading to increased resistance to 5-FU. In a cohort of healthy volunteers, H3K27me3 levels were inversely correlated with An enhancer that regulates DPYD expression was also identified. Modifying the H3K27 status of the DPYD promoter or enhancer modulates DPYD gene expression. Deletions and mutations on the DPYD promoter or enhancer abolish their transcription regulation function. Enrichment of H3K27me3 at the DPYD promoter was negatively correlated with both DPYD expression and DPD enzyme activity in peripheral blood mononuclear cellsspecimens from healthy volunteers. Lastly, tumor expression data suggests that DPYD repression by Ezh2 is a strong predictor ofs poor survival in 5-FU treated cancers. By using a combination of in silico prediction and in vitro reporter assays we also identified a functional enhancer region for DPYD that is also likely to be clinically relevant to 5-FU treatment outcome. Collectively, the presentthese findings suggest that a previously uncharacterized mechanisms regulates DPD expression and may likely contribute to tumor resistance to 5-FU. Citation Format: Rentian Wu, Qian Nie, Erin E. Tapper, Calvin R. Jerde, Steven M. Offer, Robert B. Diasio. Trimethylation and acetylation of Histone H3K27 modulates 5-fluorouracil response by regulating DPYD expression [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4069. doi:10.1158/1538-7445.AM2017-4069

Dihydropyrimidine dehydrogenase (DPD; DPYD gene) variants have emerged as reliable predictors of adverse toxicity to the chemotherapy agent 5‐fluorouracil (5‐FU). The intronic DPYD variant rs75017182 has been recently suggested to promote alternative splicing of DPYD. However, both the extent of alternative splicing and the true contribution of rs75017182 to DPD function remain unclear. In the present study we quantified alternative splicing and DPD enzyme activity in rs75017182 carriers utilizing healthy volunteer specimens from the Mayo Clinic Biobank. Although the alternatively spliced transcript was uniquely detected in rs75017182 carriers, canonically spliced DPYD levels were only reduced by 30% (P = 2.8 × 10‐6) relative to controls. Similarly, DPD enzyme function was reduced by 35% (P = 0.025). Carriers of the well‐studied toxicity‐associated variant rs67376798 displayed similar reductions in DPD activity (31% reduction). The modest effects on splicing and function suggest that rs75017182 may have clinical utility as a predictor of 5‐FU toxicity similar to rs67376798.