Automated Quantitative Analysis Scores Research Articles

Subsets of interferon signaling and response to immune checkpoint blockade.

9522 Background: Interferon (IFN) signaling in tumors is critical for both response and resistance to immune checkpoint inhibitors (ICI). While PD-L1 is an approved biomarker for ICI response in some cancers, most have no approved biomarkers for ICI response. Further, there are no clinical strategies to predict IFN sensitive vs resistant states. We hypothesized that distinct patterns of IFN signaling in melanoma, including STAT1 (precursor to IFN signaling), phosphorylated STAT1 (pSTAT1, marker of active IFN signaling), and PD-L1 (canonical inhibitory output), are associated with response or resistance to ICI. Methods: Samples from 2 tissue microarrays of 97 patients total with metastatic melanoma who received ICI from 2011-2017 were randomized into discovery [D] or validation [V] cohorts. Multiplexed immunofluorescent microscopy was used to assess STAT1, pSTAT1, and PD-L1. 54 tumors were assessed for CD3. Signals were quantified via AQUA (Automated Quantitative Analysis). High vs low signals were defined around the discovery cohort median AQUA scores. Treatment response was assessed using RECIST, and 3-year survival was analyzed. For in vitro studies, 4 human melanoma cell lines were stimulated with IFNγ/β, and Western blots were performed for STAT1, pSTAT1, and PD-L1. Results: Median follow up time was 20.2 mos (range 0.9-95.4 mos). Higher overall histospot, tumor, and nuclear STAT1 were in responders (CR/PR/SD > 6 mos) compared to non-responders (SD < 6 mos/PD, histospot p = 0.029 [D] and 0.040 [V]) and were associated with improved survival (histospot HR 0.25 [95% CI 0.08-0.76] p = 0.015 [D]; HR 0.32 [95% CI 0.12-0.84] p = 0.021 [V]). PD-L1 and pSTAT1 did not validate as associated with ICI response/survival. While many tumors had concordant high/low STAT1, pSTAT1, and PD-L1, many also had a discordant STAT1 or PD-L1 predominance. To assess the basis of this discordance, we performed IFN stimulation and Western blots of melanoma cell lines. IFNβ upregulated STAT1 greater than PD-L1 in both acute and chronic settings (p < 0.001 and < 0.001), while chronic IFNγ upregulated PD-L1 compared to the acute setting (p = 0.039) but not STAT1 (p = 0.105). Therefore, we hypothesized patients with discordant STAT1 and PD-L1 expression may have differential responses to ICI. In the overall cohort, STAT1highPD-L1high tumors (IFN engaged) were associated with improved survival compared to STAT1lowPD-L1low tumors (IFN low, HR 0.28 [95% CI 0.11-0.70] p = 0.007). Interestingly, STAT1highPD-L1low tumors were associated with improved survival compared to STAT1lowPD-L1high tumors (HR 0.28 [95% CI 0.08-0.94] p = 0.04). Accordingly, STAT1highPD-L1high and STAT1highPD-L1low tumors had greater CD3 AQUA scores compared to STAT1lowPD-L1low tumors (p < 0.001 and p = 0.018 respectively). Conclusions: Combined STAT1 and PD-L1 expression patterns may better predict melanoma response to ICI and provide insight into IFN sensitive versus resistant states.

Quantitative Analysis of Tumor-Specific BCL2 Expression in DLBCL: Refinement of Prognostic Relevance of BCL2

Introduction Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin lymphoma and a heterogeneous disease with a variety of molecular aberrations and diverse clinical outcomes. BCL2 expression has been implicated for a poor prognosis among patients with DLBCL. However, there is no consensus regarding the interpretation of BCL2 expression in DLBCL. In previous studies, the range of BCL2-positive cases, determined by immunohistochemistry (IHC), was highly variable (24 - 80%) due to subjective and semiquantitative interpretation and the absence of the established cutoff value for BCL2 expression. Consequently, the prognostic impact of BCL2 varies between studies. In addition, only the proportion of BCL2 positive tumor cells are considered in determining the BCL2 positivity. We aimed to define BCL2 positivity by quantitative analysis integrating both the intensity and proportion of BCL2 expression. Methods We retrospectively collected formalin-fixed, paraffin-embedded (FFPE) diagnostic biopsies from 221 patients with de novo DLBCL between January 2007 and December 2012 at our institute. All patients were treated with R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone) as first-line treatment. A separate validation set included 111 patients with de novo DLBCL who received first-line R-CHOP at another tertiary hospital in South Korea. BCL2 expression was analyzed using the tumor-specific Automated Quantitative Analysis (AQUA) scoring system based on multiplex immunofluorescence. The tumor-specific quantified intensity information in pixels was converted to protein expression information on a cell basis using the AQUA scoring system. The AQUA scores for BCL2 was calculated as the signal intensities of BCL2 in the target compartment divided by the pixel area of the target compartment. Cell of origin (COO) was determined according to the Hans classification using IHC. Results Ninety-eight patients (44.3%) were > 60 years old of age, 126 patients (57.0%) were male and 81 patients (36.7%) had an International Prognostic Index of 3-5 in the training set. With a median follow-up duration of 59 months, the 5-year event-free survival (EFS) and overall survival (OS) rate were 62.2% and 69.0%, respectively. The BCL2 AQUA score of 41.47 was determined as the optimal cutoff value in the ROC analysis. A total of 86 patients (38.9%) were classified as high BCL2 AQUA score group according to the determined cutoff value. High BCL2 AQUA score was significantly associated with worse OS and EFS (P = 0.00015; OS, P = 0.00012; EFS) (Figure 1A, B). Multivariate analysis revealed that high BCL2 AQUA score was a significant poor prognostic factor for both OS and EFS independent of the IPI, and COO (P = 0.01; OS, P = 0.015; EFS). The high BCL2 AQUA score group in the validation set was also significantly associated with worse OS and EFS (P = 0.0075; OS, P = 0.0049; EFS) (Figure 1C, D). The poor prognostic impact of BCL2 AQUA score was also in good correlation with both OS and EFS in the entire cohort (P < 0.0001; OS, P < 0.0001; EFS) (Figure 1E, F). The adverse impact of high BCL2 AQUA score was identified within both low (IPI 0-2) and high IPI (IPI 3-5) groups (Figure 2A-D). The high BCL2 AQUA score was also associated with poor prognosis within both GCB (P = 0.0077; OS, P = 0.0055; EFS) and non-GCB type DLBCL (P = 0.059; OS, P = 0.04; EFS), although a marginal statistical significance was observed regarding OS in the non-GCB type (Figure 2E-H). Conclusion Our study demonstrated that BCL2 expression analyzed by AQUA scoring system, incorporating both intensity and proportion, is an independent prognostic factor for patients with DLBCL. Given the growing clinical implications of BCL2 and the therapeutic progress in targeting BCL2 in hematologic malignancies, the concrete definition of BCL2 positivity in DLBCL holds great promise for the study of the pathophysiology of DLBCL and could be used to establish new therapeutic strategies. Disclosures No relevant conflicts of interest to declare.

Multiplexed quantitative analysis of CD3, CD8, and CD20 predicts response to neoadjuvant chemotherapy in breast cancer.

Although tumor-infiltrating lymphocytes (TIL) have been associated with response to neoadjuvant therapy, measurement typically is subjective, semiquantitative, and unable to differentiate among subpopulations. Here, we describe a quantitative objective method for analyzing lymphocyte subpopulations and assessing their predictive value. We developed a quantitative immunofluorescence assay to measure stromal expression of CD3, CD8, and CD20 on one slide. We validated this assay by comparison with flow cytometry on tonsil specimens and assessed predictive value in breast cancer on a neoadjuvant cohort (n = 95). Then, each marker was tested for prediction of pathologic complete response (pCR) compared with pathologist estimation of the percentage of lymphocyte infiltrate. The lymphocyte percentage and CD3, CD8, and CD20 proportions were similar between flow cytometry and quantitative immunofluorescence on tonsil specimens. Pathologist TIL count predicted pCR [P = 0.043; OR, 4.77; 95% confidence interval (CI), 1.05-21.6] despite fair interobserver reproducibility (κ = 0.393). Stromal AQUA (automated quantitative analysis) scores for CD3 (P = 0.023; OR, 2.51; 95% CI, 1.13-5.57), CD8 (P = 0.029; OR, 2.00; 95% CI, 1.08-3.72), and CD20 (P = 0.005; OR, 1.80; 95% CI, 1.19-2.72) predicted pCR in univariate analysis. CD20 AQUA score predicted pCR (P = 0.019; OR, 5.37; 95% CI, 1.32-21.8) independently of age, size, nuclear grade, nodal status, ER, PR, HER2, and Ki-67, whereas CD3, CD8, and pathologist estimation did not. We have developed and validated an objective, quantitative assay measuring TILs in breast cancer. Although this work provides analytic validity, future larger studies will be required to prove clinical utility.

136IN - How to Quantify Immuohistochemical Slides

ABSTRACT The quantitative interpretation of immunohistochemistry (IHC) is typically performed by a subjective evaluation based either on binary positive-negative or semi-quantitative assessments using intensity and percentage of positive cells. A typical example is the evaluation of HER2 expression in breast cancer, which considers also a third variable, i.e. the pattern of membrane staining. All these approaches do not provide true evaluation of the quantity of protein expression and are poorly reproducible. Assays for true protein quantification based on protein extraction, such as enzyme-linked immunosorbent assay, carry as a major drawback the loss of tissue morphology, therefore the protein expression heterogeneity cannot be evaluated. Computer-based automated analyses on classical IHC staining help eliminate the variability of pathologist-based scoring, but do not perform well in terms of quantification and are often time consuming. Some have explored the use of an automated scoring system to assess biomarker expression in tissue sections called the automated quantitative analysis (AQUA) system. This system exploits the pixel-based locale assignment for compartmentalization of expression and utilizes fluorescent tags to distinguish tumor cells from stroma, as well as to define subcellular compartments. The output variable (AQUA score) is a ‘‘pixel intensity/pixel area’’ value ranging from 0 to 255, based on the average intensity for all pixels evaluated, therefore the result is a quantitative score of immunofluorescence intensity for the tumor. It has been shown that AQUA scores are directly proportional to in situ protein concentrations. As an alternative, a technique with a high potential for better definition of protein expression is the Proximity Ligation Assay (PLA). PLA can detect single protein events such as protein protein interactions (e.g. protein dimerization) and modifications (e.g. protein phosphorylation) in tissue and cell samples prepared for microscopy. Each detected signal is visualized as an individual fluorescent dot, these signals can be quantified (counted) and assigned to a specific subcellular location based on microscopy images. Disclosure: All authors have declared no conflicts of interest.

PD-L1 Expression in Clear Cell Renal Cell Carcinoma: An Analysis of Nephrectomy and Sites of Metastases

Background: Expression of programmed death ligand (PD-L1/B7-H1/CD274) represents a mechanism of immune escape for renal cell carcinoma (RCC) cells. Drugs blocking PD-L1 or its receptor are in clinical development and early data suggests that tumor PD-L1 expression may predict response.Patients and Methods: A tissue microarray (TMA) consisting of four biopsy cores from 34 matched pairs of nephrectomy and metastatic sites of clear cell RCC was used to assess PD-L1 expression by quantitative immunofluorescence. Assessment of intra- and inter-tumor heterogeneity and primary and metastatic tumor expression was performed using a method of Automated Quantitative Analysis (AQUA).Results: The median AQUA scores were higher in metastatic than primary specimens (P < 0.0001). The correlation between PD-L1 expression in matched primary and metastatic specimens was weak (R= 0.24). Within a given tumor, variable PD-L1 staining heterogeneity was seen, however the degree of heterogeneity was similar in primary and metastatic sites (P = 0.482).Conclusions: The weak correlation between PD-L1 expression in primary and metastatic sites for a given patient suggests that expression in nephrectomy specimens cannot be used to select metastatic RCC patients for PD-L1 and PD-1 inhibitors. The intra-tumor heterogeneity seen in both primary and metastatic specimens indicates that a single core biopsy might not be sufficient to determine PD-L1 expression.

Assessment of ERCC1 and XPF Protein Expression Using Quantitative Immunohistochemistry in Nasopharyngeal Carcinoma Patients Undergoing Curative Intent Treatment

We sought to evaluate the prognostic/predictive value of ERCC1 and XPF in patients with nonmetastatic nasopharyngeal carcinoma (NPC) treated with curative intent. ERCC1 and XPF protein expression was evaluated by immunofluorescence combined with automated quantitative analysis (AQUA) using the FL297 and 3F2 antibodies, respectively. ERCC1 and XPF protein expression levels were correlated with clinical outcomes. Patient characteristics were as follows: mean age 52 years (range, 18-85 years), 67% male, 72% Karnofsky performance status (KPS) ≥ 90%, World Health Organization (WHO) type 1/2/3 = 12%/28%/60%, stage III/IV 65%. With a median follow-up time of 50 months (range, 2.9 to 120 months), the 5-year overall survival (OS) was 70.8%. Median standardized nuclear AQUA scores were used as cutpoints for ERCC1 (n=138) and XPF (n=130) protein expression. Agreement between dichotomized ERCC1 and XPF scores was high at 79.4% (kappa = 0.587, P<.001). Neither biomarker predicted locoregional recurrence, DFS, or OS after adjustment for age and KPS, irrespective of stratification by stage, WHO type, or treatment. Neither ERCC1 nor XPF, analyzed by quantitative immunohistochemistry using the FL297 and 3F2 antibodies, was prognostic or predictive in this cohort of NPC patients.

Comparative Prognostic Value of Epidermal Growth Factor Quantitative Protein Expression Compared with FISH for Head and Neck Squamous Cell Carcinoma

Epidermal growth factor receptor (EGFR) overexpression correlates with recurrence and with treatment resistance in head and neck squamous cell carcinoma (HNSCC). The aim of this study was to evaluate the relationship of EGFR gene copy number utilizing FISH and protein expression with automated quantitative analysis (AQUA) and to correlate those with patient outcome. A tissue microarray composed of 102 HNSCC treated with (chemo)radiation was constructed and analyzed for EGFR copy number by FISH (Vysis; Abbott Laboratories) and EGFR protein expression using AQUA analysis of EGFR staining scored on a scale of 0 to 255. We evaluated associations of EGFR FISH status and AQUA score with clinicopathologic parameters and survival prognosis. Eleven (17.2%) of 64 tumors with FISH results showed EGFR high polysomy and/or gene amplification (FISH positive). Protein levels assessed by AQUA in FISH-positive cases were significantly higher (P = 0.04) than in FISH-negative cases. Using the continuous AQUA scores for EGFR expression, AQUA and FISH showed significant agreement (Pearson's ρ = 0.353, P = 0.04). Patients with high tumor EGFR protein expression had inferior 5-year overall survival (27.7%) compared with those with low tumor EGFR expression (54%; P = 0.029). There was no significant association between EGFR FISH status and overall survival (P = 0.201). In the multivariate model, high tumor EGFR protein expression status remained an independent prognostic factor for overall survival (P = 0.047). EGFR protein content correlates with gene copy number if protein content is quantitated and automatically analyzed, as with AQUA. EGFR protein levels assessed by AQUA strongly predict for patient outcome in HNSCC, whereas EGFR FISH status does not provide prognostic information.

Abstract P6-04-04: Hypoxia Is Associated with Somatic Loss of BRCA1 Protein and Pathway Activity in Triple Negative Breast Cancer

Abstract Background: Triple negative breast cancer (TNBC) has loss of BRCA1 activity either through germline mutation, epigenetic modification or negative transcriptional regulators. Yale investigators have demonstrated that tumor hypoxia leads to loss of critical DNA repair activities, including BRCA1, RAD-51 and γ-H2AX (Bindra RS, et al. Cancer Res 2005;65(24): 11597-604). This study was designed to explore the relationship of BRCA1 loss and hypoxia using Carbonic Anhydrase IX (CA-IX), a downstream target of hypoxia-induced factor 1a (HIF-1a), an accepted surrogate biomarker for tumor hypoxia, in TNBC. Methods: Two cohorts of breast tumors from Yale archival materials were studied. The first consisted of 660, unselected breast tumors, the second a cohort of 130 TNBC, both with long term clinical followup. Automated Quantitative Analysis (AQUA) was used to detect the intensity of BRCA1 and CAIX within specific subcellular compartments. CA-IX antibody M75 was provided by the J Zavada lab (1:10000), BRCA1 monoclonal antibody M110 (Ab-1) was purchased from Calbiochem (1:1000). The 130 TNBC cohort was also subjected to whole genome expression analysis. In brief, tissue core biopsies from tumor blocks were subjected to nucleic acid extraction using RecoverAll Total Nucleic Acid Isolation kit (Applied Biosystems) and 600ng total RNA were processed by the Keck Microarray Facility for the Illumina DASL platform. Statistical analysis of gene expression data was carried out in Bioconductor/R software. A set of relevant signatures was selected by enrichment analysis of modules identified by principal component analysis. Signature scores were computed as Pearson correlation between the signature vector of gene contributions and each sample's expression profile for these genes. Results: In the cohort of 660 specimens, 22 were found to have high CA-IX expression score (above cut-off value of 10). Cut-off was based on known positive cell lines and corresponding AQUA scores, as well as visual confirmation of positivity. A negative correlation (Rho=-0.6, p=0.0165) of BRCA1 nuclear protein with CA-IX level was found and this data was reproducible on a duplicate array. Of note, 14 out of 22 hypoxic breast tumors were from triple negative breast cancers (TNBC) (p=0.0034). In the triple negative cohort, CAIX staining was positive in 20 % of cases (n=15/75) and in these samples it was associated with the 2002 van ‘t Veer BRCA1 mutant signature (Rho=0.51). Positive CAIX staining was also associated with worse outcome (p=0.046) as was CAIX mRNA expression (p=0.02). Conclusions: In human breast tumors BRCA1 nuclear expression is negatively correlated with tumor hypoxia, measured by CAIX. A subset of TNBC has higher levels of hypoxia and BRCA1 signatures. This data suggests that CAIX may be a useful biomarker for BRCA1 loss, and possibly for response to PARP inhibitor therapy. This will be evaluated in our recently completed trial of BSI-201+Irinotecan. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P6-04-04.

Abstract 1787: Dual targeting of PI3K and mTOR results in synergistic growth inhibition in NSCLC cells

Abstract Background: Alterations of components in the PI3K/mTOR pathway can result in malignant transformation and are found in a multitude of human cancers including non-small cell lung cancer (NSCLC). Our purpose was to characterize expression of p85 and p110α PI3K subunits in NSCLC specimens and assess the association with mTOR expression, and to study the effects of co-inhibiting PI3K and mTOR in NSCLC cell lines. Methods: Using Automated Quantitative Analysis (AQUA), a novel method of in situ protein measurement, we quantified expression of PI3K subunits in two cohorts of 251 and 192 resected NSCLC specimens. AQUA scores were correlated with clinical and pathological variables and previously published AQUA scores for mTOR. We studied changes in viability of 6 NSCLC cell lines exposed to two PI3K inhibitors, LY294002 and NVP-BKM-120, alone and in combination with rapamycin, and assessed the activity of a dual PI3K/mTOR inhibitor, NVP-BEZ-235, which is in clinical development. Results: p85 and p110α tend to be co-expressed (p&amp;lt;0.001), and higher expression of p85 was seen in adenocarcinomas than squamous cell carcinomas (SCC). High expression of p85 correlated with poor survival (p&amp;lt; 0.0075). Expression of p110α correlated with mTOR (ρ=0.30). In 6 NSCLC cell lines (3 adenocarcinomas, 3 SCC) the addition of rapamycin to LY294002 or NVP-BKM-120 was highly synergistic. Of note, even small concentrations of rapamycin (1ηM) resulted in sensitization to the PI3K inhibitors, and similar reduction in viability was seen with the addition of 1ηM, 100ηM and 1000ηM of rapamycin. NVP-BEZ-235 was highly active in NSCLC cell lines with IC50s in the ηM range and resultant down-regulation of pAkt and pP70S6K. Conclusions: PI3K is associated with decreased survival in NSCLC, suggesting that it is a good drug target for this disease. Concurrent inhibition of PI3K and mTOR is synergistic in NSCLC cell lines. Dual inhibition of PI3K and mTOR by NVP-BEZ-235 is promising and should be further evaluated in clinical trials for patients with NSCLC. 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 1787.

Abstract C21: Hypoxia-induced down-regulation of BRCA1 nuclear protein in human breast cancer tissues

Abstract Background: Poly (ADP-ribose) polymerase (PARP) inhibitors have demonstrated significant anti-tumor activity not only in tumors with BRCA1 or 2 gene mutations due to synthetic lethality, but also in triple-negative breast cancers (TNBC). The triple negative group of tumors may be more sensitive to PARP inhibitors due to loss of BRCA pathway activity, although this has not been proven. Several mechanisms may affect BRCA1 protein level such as epigenetic modification, negative transcriptional regulators, as well as hypoxia. Yale investigators have demonstrated that tumor hypoxia leads to loss of critical DNA repair activities, including BRCA1, RAD-51 and γ-H2AX (Bindra RS, et al. Cancer Res 2005;65(24):11597-604). We propose BRCA1 nuclear protein is the determinant of response to PARP inhibitors in sporadic breast cancer, a phenotype we will term ‘BRCA-like’. This study was designed to explore the relationship of BRCA1 loss and hypoxia in human breast tumors. Methods: Yale archival materials consisting of 660 human breast cancer specimens were used to establish a tissue microarray (TMA). Automated QUantitative Analysis (AQUA) was used to detect the intensity of BRCA1 and CAIXwithin a specific subcellular compartment. Carbonic Anhydrase IX (CA-IX), a downstream target of hypoxia-induced factor 1a (HIF-1a), is an accepted surrogate biomarker for tumor hypoxia. CA-IX antibody M75 was provided by the J Zavada lab (1:10000), BRCA1 monoclonal antibody M110 (Ab-1) was purchased from Calbiochem (1:1000). Results: In the cohort of 660 specimens, 22 were found to have high CA-IX expression score (above cut-off value of 10). Cut-off was based on known positive cell lines and corresponding AQUA scores, as well as visual confirmation of positivity. A negative correlation (Rho=−0.6, p=0.0165) of BRCA1 nuclear protein with CA-IX level was found in hypoxic tumors. Of note, 14 out of 22 hypoxic breast tumors were from triple negative breast cancers (TNBC) (p=0.0034). Total number of TNBC in this cohort is 112 out of 660. BRCA1 nuclear expression alone did not correlate with overall survival (p=0.76). Conclusions: In human breast tumors BRCA1 nuclear expression is negatively correlated with tumor hypoxia, measured by CAIX. Ongoing clinical trial will determine if tumors with high CA-IX or low BRCA1 benefit from PARP inhibitor therapy. Citation Information: Mol Cancer Ther 2009;8(12 Suppl):C21.

Continued high activity of figitumumab (CP-751,871) combination therapy in squamous lung cancer

8072 Background: Figitumumab (CP-751,871) is a fully human, IgG2monoclonal antibody against the insulin-like growth factor type I receptor (IGF-IR). We reported in a randomized phase II study (ASCO 2008), preliminary evidence of high activity of the combination of paclitaxel (T), carboplatin (C) and figitumumab (F) in advanced treatment-naïve NSCLC of squamous cell histology (n=11 pts). A single arm trial extension cohort was conducted to confirm those findings. Methods: Fifty-six pts with non-adenocarcinoma NSCLC were enrolled. Pts received T (200 mg/m2), C (AUC of 6) and F (20 mg/kg) q3weeks for up to 6 cycles; pts with response (PR) or stable disease were eligible to continue F as single agent until disease progression. Statistical hypotheses were 30% (null) versus 50% (response of interest). Protein expression of the IGF-IR in core tumor biopsies was quantified using automated quantitative analysis (AQUA) technology. Results: Pts were 72% male, 28% &gt;70 years old and 91% stage IV. Median number of treatment cycles was 4, with 46% of pts receiving single agent F beyond cycle 4. TCF was well tolerated. The most common all-causality grade 3, 4 CTCAE adverse events were neutropenia (21%), hyperglycemia (14%) and fatigue (14%). Hyperglycemia adverse events almost always occurred within the first treatment cycle and were manageable with standard measures. Responses in squamous pts are currently 25 out of 40 pts according to RECIST, with final response assessment still pending for 7 additional pts (≥53%, p&lt;0.001). One complete response and 7 striking PRs (50–80% tumor size reduction at cycle 2) were observed. Tumor size reductions with F maintenance treatment were also seen in 2 pts. Median progression free survival has not been yet reached at 4 months follow up. A trend (p=0.1) for higher IGF-IR expression in patients responding to TCF was observed in a small data set (n=12). Median tumor IGF-IR expression in pts responding to TCF therapy and non-responders were respectively 6287 and 4131 AQUA scores. Analysis of IGF-IR and other members of the IGF-IR pathway continues. Conclusions: These data further support the activity of figitumumab combination therapy in pts with squamous NSCLC. [Table: see text]

A comprehensive evaluation of the prognostic value of ERCC1 protein expression, mRNA and genotype in locally advanced squamous cell carcinoma of the head and neck (LA-SCCHN)

6011 Background: ERCC1 has been associated with resistance to cisplatin (CDDP). In this retrospective study we analyzed ERCC1 protein expression, mRNA levels and single nucleotide polymorphisms (SNP) at codon 118 and examined each as a potential predictive/prognostic marker in LA-SCCHN pts treated with concurrent CDDP and radiation (CCRT). Methods: Records of 91 consecutive LA-SCCHN pts treated at the Tom Baker Cancer Centre between Aug 2000 and Mar 2005 were reviewed. ERCC1 was evaluated from pre-treatment formalin-fixed paraffin-embedded (FFPE) biopsies. Protein expression was assessed by conventional immunohistochemistry (IHC) using a 4-point semiquantitative scale as well as by immunofluorescence combined with automated quantitative analysis (AQUA). Quantitative RT-PCR and a pre-made TAQMAN genotyping assay were used to evaluate mRNA levels and the codon 118 SNP respectively. Results: FFPE biopsies were available from 57 pts (63%). Patient characteristics were as follows: mean age 56.4 years (range 43–75), 82.5% male, 77.2% KPS > 80%, 50.9% oropharyngeal primaries. ERCC1 protein expression was high in 39 pts (72%) by IHC versus 17 pts (31%) using AQUA. Ten pts (26%) had high levels of mRNA. C/C, C/T and T/T genotype frequencies were 17%, 43%, and 40%, respectively. ERCC1 IHC grading was significantly Spearman correlated (r=0.47) with AQUA scores. mRNA and protein expression were not correlated. The complete response rate was 89.3%. With a median follow-up of 44 months (range 3, 85), the 5 year overall survival (OS) was 68.2%. ERCC1 levels did not predict response, regardless if the pts received 100% versus <100% of CDDP. A nonsignificant increase in locoregional recurrence (LRR), recurrence at any site (ALLREC) and worse OS was observed with high IHC score. The C/C and C/T genotype was also predictive of LRR (HR=7.4, 95%CI: 0.9–58.6, p=0.06), ALLREC (HR=8.6, 95%CI: 1.1–67.2, p=0.04), and higher mortality (HR=3.9, 95%CI: 0.8–17.8, p=0.08). mRNA levels did not predict outcome. Conclusions: Our preliminary results suggest that high ERCC1 protein expression as well as the C/C or C/T genotype may be potential prognostic markers for LA-SCCHN pts treated with CCRT. No significant financial relationships to disclose.

Assessment of PI3 kinase as a druggable target in melanoma

8521 Background: PI3 (phosphatidylinositol-3) kinase is central to a major intracellular signal transduction pathway that influences numerous cellular functions, including growth, differentiation and survival. Drugs that target PI3 kinase have demonstrated preclinical activity, and are undergoing further development. The PI3 kinase inhibitor, SF1126 is expected to enter clinical trials in 2007. The hallmarks for a valuable drug target include differential expression between normal and malignant cells and an association with disease progression. PI3 kinase expression has not been thoroughly evaluated in melanoma. We sought to determine the expression and prognostic value of PI3 kinase in a large cohort of primary and metastatic melanomas and to compare the expression with that of benign nevi. Methods: Tissue microarrays containing 548 melanomas and 540 benign nevi were employed to assess PI3 kinase expression. We used a novel method of in situ automated quantitative analysis (AQUA) of protein levels. We applied S100 conjugated to Cy3 to identify a melanoma tumor mask within a histospot, and measured PI3 kinase expression levels using Cy5 conjugated antibodies within the mask. Continuous AQUA scores were correlated with clinical and pathological variables. Results: PI3 kinase expression was higher in melanomas than in nevi by unpaired t-tests (p &lt; 0.0001), and was significantly higher in metastatic than in primary specimens (p &lt; 0.0001). High PI3 kinase scores were associated with Clark levels of IV-V (p = 0.0126), Breslow thickness &gt; 2mm (p = 0.044) and absence of tumor infiltrating lymphocytes (p = 0.014). High PI3 kinase expression strongly correlated with decreased survival by Cox univariate analysis (p = 0.0024), but was not independent of disease stage. Conclusions: PI3 kinase expression is remarkably higher in malignant melanocytes than in benign nevi. Moreover, high PI3 kinase expression is associated with disease aggression, making PI3 kinase an attractive drug target for melanoma. This is the first large study assessing PI3 kinase levels in melanoma in a quantitative fashion. Assessment of the association between PI3 kinase levels and response to therapy is warranted. No significant financial relationships to disclose.

AQUA and FISH analysis of HER‐2/neu expression and amplification in a small cell lung carcinoma tissue microarray

Most small cell lung carcinoma (SCLC) patients have metastatic disease at the time of diagnosis and are faced with poor prognosis and limited treatment options. Reports of HER-2/neu gene amplification and overexpression in this malignancy have raised the possibility of applying targeted immunotherapy with trastuzumab, the monoclonal antibody used to treat metastatic breast cancer. However, a review of the studies measuring HER-2/neu gene amplification and protein expression in SCLC reveals discordant results. The aim of the present study was to re-examine HER-2/neu expression in SCLC in relation to gene copy number using the new, highly sensitive, immunofluorescence automated quantitative analysis (AQUA) technology. Fluorescence in situ hybridization (FISH) was used to measure HER-2/neu gene copy number and amplification status and AQUA was used to measure protein expression in a series of 23 SCLC tumours on a tissue microarray. None of the 17 SCLC specimens assessable by FISH exhibited HER-2/neu gene amplification as defined by a HER-2/neu/chromosome 17 ratio = or > 2. Twelve of 17 (70.1%) SCLC samples were polysomic for chromosome 17 with corresponding increases in HER-2/neu gene copy numbers. Intermediate levels of protein expression corresponding to AQUA scores in the range of 4-24 were detected in all 23 specimens. High protein expression levels corresponding to AQUA scores up to 83, observed previously in association with gene amplification and poor prognosis in breast cancer cases, were not detected in the present study. No statistically significant association was observed between absolute chromosome 17 or HER-2/neu gene copy numbers and protein expression levels in tumour cells (P > 0.45). The lack of gene amplification and robust HER-2/neu protein expression in SCLC tumour cells in this series does not suggest a prominent role for the HER-2/neu gene in SCLC tumour progression and does not support the general applicability of targeted immunotherapy with trastuzumab to this malignancy.

Evaluating the Expression and Prognostic Value of TRAIL-R1 and TRAIL-R2 in Breast Cancer

The cell surface receptors tumor necrosis factor-related apoptosis-inducing ligand receptor 1 (TRAIL-R1) and TRAIL-R2 transmit apoptotic signals, and agents that activate these receptors are in clinical development. We sought to determine the expression and prognostic value of TRAIL-R1 and TRAIL-R2 in early-stage breast cancer. Tissue microarrays containing specimens from 655 breast cancer patients with 20-year follow-up were employed and evaluated with our automated quantitative analysis (AQUA) system. The system uses cytokeratin to define pixels as breast cancer (tumor mask) within the array spot, and measures intensity of TRAIL receptor expression using Cy5 conjugated antibodies within the mask. AQUA scores were correlated with clinical and pathologic variables. TRAIL-R1 and TRAIL-R2 expression were similarly studied on 95 unmatched normal breast specimens. TRAIL-R1 expression was not associated with survival. High TRAIL-R2 expression strongly correlated with decreased survival (P = 0.0005). On multivariate analysis, high TRAIL-R2 expression remained an independent prognostic marker, as did nodal status and tumor size. High TRAIL-R2 expression correlated strongly with lymph node involvement (P = 0.0003). TRAIL-R2 expression was stronger in malignant specimens than in normal breast epithelium (P < 0.0001). High TRAIL-R2 expression was independently associated with decreased survival in breast cancer. The biological basis and the sensitivity of high TRAIL-R2 expressing cells to TRAIL agonists and/or chemotherapy are subject to further investigation. Evaluation of TRAIL-R2 expression in early-stage breast cancer may identify a subset of patients requiring more aggressive or pathway-targeted adjuvant treatment. Clinical trials involving TRAIL-R2 agonists should stratify patients based on TRAIL-R2 expression.