Urine based assays that can non-invasively detect bladder cancer (BCa) have the potential to reduce unnecessary and invasive procedures. The purpose of this study was to develop a multiplex immunoassay that can accurately and simultaneously monitor ten diagnostic urinary protein biomarkers for application as a non-invasive test for BCa detection. A custom electrochemiluminescent multiplex assay was constructed (Meso Scale Diagnostics, LLC, Rockville, MD, USA) to detect the following urinary proteins; IL8, MMP9, MMP10, ANG, APOE, SDC1, A1AT, PAI1, CA9 and VEGFA. Voided urine samples from two cohorts were collected prior to cystoscopy and samples were analyzed blinded to the clinical status of the participants. Means (±SD) and receiver operating characteristic (ROC) curve analysis were used to compare assay performance and to assess the diagnostic accuracy of the diagnostic signature. Comparative diagnostic performance analyses revealed an AUROC value of 0.9258 for the multiplex assay and 0.9467 for the combination of the single-target ELISA assays (p = 0.625), so there was no loss of diagnostic utility for the MSD multiplex assay. Analysis of the independent 200-sample cohort using the multiplex assay achieved an overall diagnostic sensitivity of 0.85, specificity of 0.81, positive predictive value 0.82 and negative predictive value 0.84. It is technically feasible to simultaneously monitor complex urinary diagnostic signatures in a single assay without loss of performance. The described protein-based assay has the potential to be developed for the non-invasive detection of BCa.

We set out to determine if the administration of subcutaneous (SQ) ALT-803 was non-inferior to standard intravesical BCG treatment in a carcinogen induced mouse (C57BL/6J) bladder cancer model. Using this well-established carcinogen induced mouse model, we studied the effects of various dosing schemas of ALT-803 (SQ alone, SQ with intravesical BCG, intravesical alone, intravesical with intravesical BCG) compared to intravesical BCG alone (positive control) and PBS (negative control). The non-inferiority margin for the difference in bladder weight, as a surrogate for tumor mass, was defined as 7%. All treatment groups (i.e., ALT-803 SQ alone, ALT-803 SQ with intravesical BCG, ALT-803 intravesical alone, ALT-803 intravesical with intravesical BCG and intravesical BCG alone) demonstrated a significant reduction in tumor burden as evident by bladder weights and H&E stain (p < 0.005). Non-inferiority tests between the intravesical BCG alone group and the additional treatment groups showed that SQ ALT-803 alone (p = 0.04) and BCG plus SQ ALT-803 (p = 0.009) were non-inferior to intravesical BCG alone. In this model, we did not see an appreciable infiltration of CD4+ T, CD8+ T or CD161/KLRB1+ natural killer (NK) cells in the bladder/tumor. When assessing peripheral blood mononuclear cells, SQ ALT-803 alone resulted in a robust induction of CD8+ T cells (p < 0.01), NKG2D+ NK cells (p < 0.005) and CD3+/NKG2D+ NKT cells (p < 0.005) compared to other groups, while in splenic tissue, SQ ALT-803 alone resulted in a robust induction of CD3+/NKG2D+ NKT cells (p < 0.005) compared to other groups. Subcutaneous ALT-803 treatment alone or in combination with intravesical BCG was well tolerated and was not inferior to intravesical BCG alone. CD8+ T, NKG2D+ NK and CD3+/NKG2D+ NKT cell induction along with induction of key cytokines remain steadfast mechanisms behind ALT-803. The enhanced therapeutic index seen with BCG and ALT-803, administered SQ or intravesically, provides a powerful justification for the further development of these regimens.

Accumulating evidence suggests that sphingosine kinase 1 (SphK1)/sphingosine 1-phosphate pathway plays a pivotal role in colon carcinogenesis. To further support the evidence, we investigated the effects of SphK1 using three separate animal models: SphK1 knockout mice, SphK1 overexpressing transgenic mice, and SphK1 overexpression in human colon cancer xenografts. Using azoxymethane (AOM, colon carcinogen), we analyzed colon tumor development in SphK1 KO and SphK1 overexpression in intestinal epithelial cells regulated by a tet-on system. Then, we analyzed subcutaneous tumor growth using xenografts of HT-29 human colon cancer cell. Finally, immunohistochemical analyses for SphK1 and COX-2 were performed on human colon cancer tissue microarray. SphK1 KO mice, compared to wild-type mice, demonstrated a significant inhibition in colon cancer development induced by AOM (58.6% vs. 96.4%, respectively, P < 0.005). Tumor multiplicity (1.00 vs. 1.64 per colon, respectively, P < 0.05) and tumor volume (14.82 mm3 vs. 29.10 mm3, P < 0.05) were both significantly reduced in SphK1 KO mice compared to wild-type mice. Next, SphK1 overexpression in HT-29 enhanced tumor growth as compared to GFP control in nude mice (229.5 mm3 vs. 90.9 mm3, respectively, P < 0.05). Furthermore, overexpression of SphK1 in intestinal epithelial cells significantly enhances AOM-induced colon tumor formation (P < 0.05). Lastly, SphK1 and COX-2 intensity tended to reduce overall survival of late stage colon cancer patients. SphK1 expression regulates the early stage of colon carcinogenesis and tumor growth, thus inhibition of SphK1 may be an effective strategy for colon cancer chemoprevention.

We set out to expand on our previous work in which we reported the epithelial expression pattern of a urine-based bladder cancer-associated diagnostic panel (A1AT, ANG, APOE, CA9, IL8, MMP9, MMP10, PAI1, SDC1, and VEGFA). Since many of the analytes in the bladder cancer-associated diagnostic signature were chemokines, cytokines, or secreted proteins, we set out to report the stromal staining pattern of the diagnostic signature as well as CD3⁺ (T-cell) cell and CD68⁺ (macrophage) cell staining in human bladder tumors as a snapshot of the tumor immune landscape. Immunohistochemical staining was performed on 213 tumor specimens and 74 benign controls. Images were digitally captured and quantitated using Aperio (Vista, CA). The expression patterns were correlated with tumor grade, tumor stage, and outcome measures. We noted a positive correlation of seven of the 10 proteins (excluding A1AT and IL8 which had a negative association and VEGFA had no association) in bladder cancer. The overexpression of MMP10 was associated with higher grade disease, while overexpression of MMP10, PAI1, SDC1 and ANG were associated with high stage bladder cancer and CA9 was associated with low stage bladder cancer. Increased tumor infiltration of CD68⁺ cells were associated with higher stage disease. Overall survival was significantly reduced in bladder cancer patients’ whose tumors expressed eight or more of the 10 proteins that comprise the bladder cancer diagnostic panel. These findings confirm that the chemokines, cytokines, and secreted proteins in a urine-based diagnostic panel are atypically expressed, not only in the epithelial component of bladder tumors, but also in the stromal component of bladder tumors and portends a worse overall survival. Thus, when assessing immunohistochemical staining, it is important to report staining patterns within the stroma as well as the entire stroma itself.

Mammographic density is a known risk factor for breast cancer, but the underlying pathologic characteristics are not well understood. The current analysis investigated the expression of several markers of interest, e.g., inflammation and growth, with mammographic density (MD) in normal and malignant breast tissue specimens from 279 women of the Multiethnic Cohort (MEC). Breast cancer cases, recruited from a nested case-control study within the MEC, provided mammograms for density evaluation. Protein expression (COX-2, TNF-α, TGF-β, IGF-1R, IGFBP-2, and vimentin) was assessed by immunohistochemical detection. Linear regression was applied to evaluate the relation between marker expression and percent density and to compute adjusted means with 95% confidence intervals (CI) by marker status while adjusting for confounders. Due to missing cores and tissue, normal tissue could only be evaluated for COX-2 and vimentin. No significant associations with mammographic density were detected for all markers analyzed. For inflammatory markers (TNF-α, COX-2, and TGF-β) in tumor tissue, MD were non-significantly higher with stronger expression but the differences were very small. For example, the mean MD values for no, weak, and strong TNF-α expression were 35% (95% CI 24–47%), 39% (95% CI 29–48%), and 38% (95% CI 27–50%). In a posthoc analysis among postmenopausal women only, the difference across categories of TNF-α expression increased to 25% (95% CI 12–39%), 35% (95% CI 23–48%), and 35% (95% CI 20–49%). The current analysis offers little support for an involvement of immunohistochemical markers representing inflammatory and growth factor pathways as predictors of breast density.

Technological advances enable the cost-effective acquisition of Multi-Modal Data Sets (MMDS) composed of measurements for multiple, high-dimensional data types obtained from a common set of bio-samples. The joint analysis of the data matrices associated with the different data types of a MMDS should provide a more focused view of the biology underlying complex diseases such as cancer that would not be apparent from the analysis of a single data type alone. As multi-modal data rapidly accumulate in research laboratories and public databases such as The Cancer Genome Atlas (TCGA), the translation of such data into clinically actionable knowledge has been slowed by the lack of computational tools capable of analyzing MMDSs. Here, we describe the Joint Analysis of Many Matrices by ITeration (JAMMIT) algorithm that jointly analyzes the data matrices of a MMDS using sparse matrix approximations of rank-1. The JAMMIT algorithm jointly approximates an arbitrary number of data matrices by rank-1 outer-products composed of “sparse” left-singular vectors (eigen-arrays) that are unique to each matrix and a right-singular vector (eigen-signal) that is common to all the matrices. The non-zero coefficients of the eigen-arrays identify small subsets of variables for each data type (i.e., signatures) that in aggregate, or individually, best explain a dominant eigen-signal defined on the columns of the data matrices. The approximation is specified by a single “sparsity” parameter that is selected based on false discovery rate estimated by permutation testing. Multiple signals of interest in a given MDDS are sequentially detected and modeled by iterating JAMMIT on “residual” data matrices that result from a given sparse approximation. We show that JAMMIT outperforms other joint analysis algorithms in the detection of multiple signatures embedded in simulated MDDS. On real multimodal data for ovarian and liver cancer we show that JAMMIT identified multi-modal signatures that were clinically informative and enriched for cancer-related biology. Sparse matrix approximations of rank-1 provide a simple yet effective means of jointly reducing multiple, big data types to a small subset of variables that characterize important clinical and/or biological attributes of the bio-samples from which the data were acquired.