Anthony Sferruzza

Natural history of celiac disease autoimmunity in a USA cohort followed since 1974

Abstract Background. The natural history and the possible changes of celiac disease (CD) prevalence over time are still unclear. Objectives. 1) To establish whether loss of tolerance to gluten may occur at any age; 2) to investigate possible changes of CD prevalence over time; and 3) to investigate CD-related co-morbidities. Methods. We analyzed 3,511 subjects with matched samples from 1974 (CLUE I) and 1989 (CLUE II). To avoid a selection bias regarding survival, we also screened 840 CLUE I participants who deceased after the 1974 survey. Outcome measure. CD autoimmunity (positivity to auto-antibodies) over time. Results. CD autoimmunity was detected in seven subjects in 1974 (prevalence 1:501) and in an additional nine subjects in 1989 (prevalence 1:219). Two cases of CD autoimmunity were found among the 840 subjects deceased after CLUE I. Compared to controls, untreated CD subjects showed increased incidence of osteoporosis and associated autoimmune disorders, but they did not reach statistical significance. Conclusions. During a 15-year period CD prevalence increased 2-fold in the CLUE cohort and 5-fold overall in the US since 1974. The CLUE study demonstrated that this increase was due to an increasing number of subjects that lost the immunological tolerance to gluten in their adulthood.

Distribution of the ugt1a1*28 polymorphism in Caucasian and Asian populations in the Us: a genomic analysis of 138 healthy individuals

The hepatic isoform 1A1 of uridine diphosphate glucuronosyltransferase is responsible for glucuronidation and detoxification of SN-38, the active metabolite of irinotecan. The presence of an additional TA repeat in the TATA sequence of the UGT1A1 promoter leads to a significant decrease in SN-38 glucuronidation. Patients with the UGT1A1 (TA)7 allele are more likely to experience severe neutropenia and diarrhea following irinotecan chemotherapy. We assessed the distribution of the UGT1A1 (TA)n polymorphism in healthy male and female US residents of European and Asian descent. We used a fluorescent polymerase chain reaction-based assay to detect UGT1A1 (TA)n polymorphisms in 138 healthy volunteers (56 Caucasians, 37 Chinese, 37 Filipino and eight Japanese) between the ages of 18 and 65 years. The χ2-test was used to assess between-group differences in the distribution of UGT1A1 (TA)n genotypes. The UGT1A1 (TA)6/6 genotype was significantly more common in Asians than in Caucasians (76 vs. 46%), whereas the (TA)6/7 (39 vs. 20%) and (TA)7/7 (13 vs. 5%) genotypes were more common in Caucasians than in Asians. Genotype distributions did not differ significantly between men and women in either group. The UGT1A1 (TA)5/5 genotype was detected in one Caucasian woman. In conclusion, consistent with previous reports, the UGT1A1 (TA)7/7 genotype was significantly more common in Caucasians than in Asians. UGT1A1 (TA)n/n genotype distribution did not vary with sex in individuals of European or Asian descent.

Measurement of von Willebrand factor‐FVIII binding activity in patients with suspected von Willebrand disease type 2N: application of an ELISA‐based assay in a reference laboratory

Summary.  Laboratory diagnosis of von Willebrand disease type 2N (VWD2N) is based on costly mutation analysis or in vitro measurement of the ability of plasma von Willebrand factor (VWF) to bind exogenous factor VIII (FVIII); however, the VWF‐FVIII binding activity assay is complex and not widely used. Our aim was to assess the utility of the in‐house VWF‐FVIII binding assay in the investigation of patients with suspected VWD2N. A previously described ELISA‐based FVIII binding method was simplified and adapted for the clinical laboratory use by optimizing incubation time, reagent concentrations and assay standardization. The assay was validated using samples from eight individuals with known homozygous or heterozygous VWD2N mutations, and 100 healthy adults. An additional 392 patient samples were tested, including 314 with FVIII activity <50% of normal and 78 received for routine VWF‐FVIII binding activity testing. Intra‐ and inter‐assay variations were less than 10% and 17%, respectively, and the limit of quantification was estimated as 0.12. The reference range for healthy adults was 0.73–1.42. VWF:FVIII binding activity was consistent with the genotype in subjects with available genetic data, being low in three individuals with homozygous mutation (<0.12) and intermediate in five heterozygous individuals (0.44–0.61). Screening of the 392 clinical samples identified reduced VWF:FVIII binding in 19 subjects. This assay provides accurate measurement of VWF:FVIII binding activity and successfully identifies homozygous VWD2N patients and heterozygous carriers. Use of this ELISA‐based assay may help avoid the need for mutation analysis in patients with unexplained low FVIII activity.

Abstract 4675: Detection of ALK, ROS1, and RET translocations in non-small cell lung cancer (NSCLC) patients by intragenic differential expression analysis

BACKGROUND: ALK, ROS1, and RET translocations are frequently detected in NSCLC patients. Crizotinib, a tyrosine kinase inhibitor (TKI), was approved by the FDA in 2011 to treat NSCLC in patients harboring ALK translocations as detected by an FDA-approved assay. However, the FDA-approved ALK FISH assay is technically challenging, with failures due to pre-analytic variables. Another approach, intragenic differential expression (IDE), detects translocations by comparing expression levels of the 5′ end with the 3′ end of target gene transcripts. In this study we developed and evaluated a rapid IDE assay to screen for ALK, ROS1, and RET translocations, independent of the fusion partner. METHODS: A total of 419 samples (408 randomly-selected NSCLC clinical samples, ALK positive and ROS1 positive cell lines (2 each), and 7 previously-tested RET-positive clinical samples) were used to develop and evaluate performance characteristics of the IDE assays. To determine IDE scores, levels of ALK, ROS1, and RET expression were first determined by quantitative RT-PCR measurement of the 5′- and 3′- ends of the respective transcripts. The differences in expression levels were calculated as ΔCt (Ct5′ - Ct3′). High ΔCt values indicate presumptive presence of gene translocations. 212/408 NSCLC samples were analyzed by ALK FISH and EML4-ALK RT-PCR, and 196/408 samples were analyzed by EML4-ALK RT-PCR. RESULTS: Thirty-one of the 408 (7.6%) clinical samples tested positive for ALK rearrangements by IDE. Among them, 20 were confirmed by FISH and/or EML4-ALK (true positive, 64.5%), while 11 were negative by FISH and/or EML4-ALK (false positive, 35.5%). One of 10 ALK FISH positive samples tested negative by both ALK IDE and EML4-ALK RT-PCR analysis (false negative), while one of 202 FISH-negative sample tested positive by both EML4-ALK and ALK IDE. ALK IDE exhibited 94.5% (189/200) concordance with ALK FISH and 96.0% (356/371) concordance with the EML4-ALK assay. For ROS1, both ROS1-positive cell lines and 4/408 (1.0%) NSCLC samples tested positive for ROS1 by IDE. Among the 4 IDE-positive NSCLC samples, 1 was confirmed by ROS1 FISH. For RET, all 7 known positives and 10/408 (2.5%) NSCLC samples tested positive by IDE. Three of six RET IDE positive NSCLC samples were confirmed by RET FISH. Overall, ALK, ROS1, and RET translocations were mutually exclusive in NSCLC patients. The lung IDE assay had a failure rate of 3.7%. CONCLUSION: These findings demonstrate the feasibility of using IDE to detect ALK, ROS1, and RET gene translocations. These assays may have potential as a screening tool to select patients for further confirmation by FISH for TKI-targeted therapy. The IDE concept can be applied to a wide range of somatic translocations. Citation Format: Shih-Min Cheng, Cindy Barlan, Feras Hantash, Heather R. Sanders, Patricia H. Chan, Vladimira Sulcova, Marc A. Sanidad, Kevin Qu, Joann C. Kelly, Fatih Z. Boyar, Anthony D. Sferruzza, Frederic M. Waldman. Detection of ALK, ROS1, and RET translocations in non-small cell lung cancer (NSCLC) patients by intragenic differential expression analysis. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4675. doi:10.1158/1538-7445.AM2014-4675

Abstract 4569: Amplicon-targeted ultra-deep sequencing for the detection of KRAS, BRAF, and PIK3CA actionable mutations

Proceedings: AACR 103rd Annual Meeting 2012‐‐ Mar 31‐Apr 4, 2012; Chicago, IL Background: Next-generation deep sequencing techniques have the potential to revolutionize clinical testing, especially in the arena of mutation detection. Molecular diagnostic testing for KRAS, BRAF, and PIK3CA mutations has become an integral part of clinical practice for the determination of treatment eligibility for anti-EGFR therapy. Purpose: Here we investigate the feasibility of ultra-deep advanced sequencing with the 454 GS Junior instrument (454 Life Sciences/Roche, Branford CT) for the simultaneous detection of actionable mutations in KRAS, BRAF, and PIK3CA in formalin-fixed paraffin-embedded (FFPE) colorectal cancer tissue. Methods: Advanced sequencing with a 4-amplicon panel was performed on 17 de-identified residual FFPE colorectal cancer biopsy specimens originally submitted for mutation analysis of one or more genes by standard sequencing. The 4-amplicon panel was multiplexed for up to 4 samples through “barcoding” with molecular identifier (MID) sequences, for a total of 16 amplicons per run. Results: Sequence analysis with ultra-deep coverage of 3000-5000 reads detected KRAS mutations in 5 (29%) tumor specimens, BRAF mutations in 3 (18%), and PIK3CA mutations in 4 (24%). Ten cases had a single mutation and one had two: PIK3CA E545K and BRAF D594G. Coverage was equivalent for each of the four amplicons, and in both the forward and reverse directions, providing a high level of confidence for each mutation detected. Sequencing data was obtained for each sample tested. When sample quantity allowed, Sanger sequencing and/or castPCR (Life Technologies, Grand Island, NY) were performed as method comparisons, and results were concordant. Conclusions: The use of MIDs with ultra-deep advanced sequencing allowed the simultaneous detection of mutations in multiple amplicons from multiple patients in a single sequencing run. This proof-of-concept study confirmed that using advanced sequencing to simultaneously test for multiple biomarkers in one FFPE sample is feasible, and that amplicon-based advanced sequencing is a robust and high-throughput platform for the detection of clinically actionable mutations. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 4569. doi:1538-7445.AM2012-4569

Abstract 3156: MicroRNA profiling for the detection of melanoma

Background: Numerous reports have suggested a link between aberrant expression of microRNAs (miRNAs) and various forms of cancer, including melanoma. We examined miRNA expression patterns in melanoma and nevi to identify candidate biomarkers for differentiating between these melanocytic lesions. Method: The study used formalin-fixed paraffin-embedded (FFPE) tissues from 78 patients with melanoma and 98 with nevi (18 compound nevi, 20 intradermal nevi, 20 junctional nevi, 20 blue nevi, and 20 spitz nevi). All lesions were confirmed by at least 2 pathologists. Total RNA was extracted from the FFPE tissue and hybridized onto the Affymetrix GeneChip miRNA array (Affymetrix, Santa Clara, CA), which measure 850 human microRNAs. After determining an optimal set of miRNAs for differentiation using random forest algorithm, we retested specimens for this set using quantitative real-time PCR (qPCR) on a TaqMan miRNA system (Applied Biosystems, Foster City, CA). Results: Expression levels of 257 miRNAs differed significantly between melanomas and nevi (p values 0.75 for distinguishing melanomas from nevi, and 74 exhibited more than a two-fold difference. An optimized set of 7 miRNAs provided good differentiation of melanomas from nevi, with an AUORC of 0.94 and 87.7% concordance with pathology-determined tissue type. miRNA array data were further validated by qPCR: for 5 of the 7 miRNAs identified above, qPCR results showed significant correlation with the miRNA array results (average Pearson correlation coefficient = 0.95). Conclusions: These data suggest that the miRNA expression pattern from FFPE tissue holds promise for distinguishing melanomas from nevi. Use of this miRNA set as a diagnostic marker requires further validation and exploration in a large-scale study. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 3156. doi:1538-7445.AM2012-3156

Abstract 3728: Two novel EML4-ALK fusion variants involving EML4 exon 17 identified in lung cancer

Proceedings: AACR 101st Annual Meeting 2010‐‐ Apr 17‐21, 2010; Washington, DC Introduction: Recently, anaplastic lymphoma kinase (ALK) inhibitors have been used successfully in patients harboring gene fusions between echinoderm microtubule-associated protein-like 4 (EML4) and ALK. These fusions result from a paracentric inversion on chromosome 2 inversion [inv(2)(p21;p23)] and have been identified in 3-7% of all non-small cell lung cancer (NSCLC) cases. To date, 11 variants have been published involving 7 different EML4 exons and invariably involving exon 20 of ALK. To screen for EML4-ALK abnormalities, we developed a multiplex RT-PCR exon screening approach that can detect variants initiating at any of the first 22 EML4 exons. Using this approach we detected both known and novel EML4-ALK fusion variants. We identified 2 novel EML4-ALK fusion variants involving exon 17 of EML4 from a single patient. Methods: Sixty-one samples were analyzed with the multiplex assay: 56 formalin-fixed paraffin-embedded (FFPE) NSCLC samples, 4 cell line samples (3 NSCLCs; 1 Prostate carcinoma), and 1 control RNA (Human Total RNA, Applied Biosystems). After RNA extraction, RT-PCR was performed using the RNA UltraSense™ One-Step qRT-PCR System (Invitrogen). Twenty-three primers (22 unlabeled EML4 forward; 1 FAM- labeled ALK reverse) were included in 4 master mixes to amplify EML4-ALK fusions initiating within the first 22 EML4 exons to ALK exon 20; 1 endogenous control (beta-2-microglobulin) primer set was included in a separate reaction. After RT-PCR, the PCR products were separated by capillary electrophoresis on a genetic analyzer (ABI 3730, Applied Biosystems) and the fusions identified based on size (bp). Samples with positive results were further analyzed by singleplex RT-PCR to confirm exon involvement and novel fusions were confirmed by sequencing. Results: One NSCLC cell line was positive for EML4-ALK variant 3a and 3b and 2 NSCLC cell lines were negative, consistent with literature findings. The prostate cancer and control RNA were negative. Two unexpected peaks resulted from multiplex and singleplex RT-PCR reactions containing EML4 exon 17 forward and ALK exon 20 reverse primers. Sequencing revealed 2 previously undescribed variants. One variant (8a) consisted of a complete EML4 exon 17 fused to a partial intron 19-20 and complete exon 20 of ALK. The second variant (8b) consisted of a complete exon 17 with partial intron 17-18 of EML4 fused to the same ALK region as in variant 8a. Overall, 9% (5/56) of lung cancer tumor tissue (FFPE) were positive for EML4-ALK fusions (variants 3a and 3b; 2/56); variant 3a only (1/56), variant 1 (1/56), and novel variants 8a and 8b (1/56)). Conclusions: This study demonstrates that proper testing for EML4-ALK fusion in lung cancer should encompass at least most of the first 22 exons of the EML4 gene. Encompassing multiplex RT-PCR assays may increase the detection prevalence of EML4-ALK fusion with the detection of new fusion variants as demonstrated in our small series of patients. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 3728.

Abstract 3745: Detection of various ALK translocations using intragenic differential expression (IDE) in patients with non-small cell lung cancer

Proceedings: AACR 101st Annual Meeting 2010‐‐ Apr 17‐21, 2010; Washington, DC Introduction: Efforts to target EML4-ALK fusions with ALK inhibitors in non-small cell lung cancer (NSCLC) have shown promising results in patients harboring a paracentric inversion on chromosome 2, inv(2)(p21p23). In effect, the clinical utility of these drugs is dependent on the presence of ALK gene activation. Reliable testing for ALK activation by translocation is important for selecting patients for this therapy. Although 11 variants for translocation are known, new variants have recently been reported. The aim of this study was to develop a reliable molecular assay to detect translocation of the ALK gene irrespective of the breakpoint on the EML4 gene or the partner gene. To achieve this, we used intragenic differential expression (IDE) of ALK gene comparing expression levels of the 5′ end with the 3′ end. Methods: ALK IDE was determined by measuring the levels of both 5′ and 3′ transcript regions by quantitative RT-PCR. IDE scores were obtained by calculating the endogenous control (ABL1) normalized differences in 5′ and 3′ ALK levels (IDE = 5′ALK/ABL1 - 3′ALK/ABL1). High IDE score, relative to normal, indicates the presence of ALK rearrangement. Relative expression of ALK (independent of rearrangement) was also established. A subset of samples were also analyzed by fluorescence in situ hybridization (FISH). All study samples were analyzed for direct detection of EML4-ALK by RT-PCR performed in a parallel study. Results: The ALK IDE value of an EML4-ALK fusion-positive cell line (NCI-H2228) was 0.7 whereas it was 0.0 for 2 EML4-ALK-negative NSCLC cell lines (NCI H838 and NCI H1299). The positive control value (0.7) was set as the cutoff to distinguish positive vs. negative ALK rearrangement. Eleven percent (6/56) of the lung cancer tissue samples were IDE positive. Eighty-three percent (5/6) of these positives were confirmed by direct detection of EML4-ALK fusion transcript by RT-PCR, including one specimen harboring the previously undescribed variants 8a and 8b. Five IDE-positive samples and 5 with slightly-to-moderately elevated levels of ALK transcript (3′ ALK > 0.1) were further analyzed by FISH. Of these 10 samples, 80% (8/10) showed ALK rearrangement and/or gene amplification. All samples interpreted as having ALK rearrangements by FISH were also positive by IDE (3/3). Two IDE positive (1 confirmed, 1 unconfirmed by RT-PCR) were interpreted as rearrangement negative by FISH. Conclusions: ALK IDE accurately categorized all FISH-confirmed rearrangements as positive and detected rearrangements in at least 1 other confirmed case not identified by FISH. This method is useful for detection of EML4-ALK rearrangements and may function as a universal molecular assay for determining ALK rearrangements in multiple tumor types. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 3745.