Overcoming intratumoural heterogeneity for reproducible molecular risk stratification: a case study in advanced kidney cancer

Artificial Intelligence in Renal Cell Carcinoma Histopathology: Current Applications and Future Perspectives

Renal cell carcinoma (RCC) is characterized by its diverse histopathological features, which pose possible challenges to accurate diagnosis and prognosis. A comprehensive literature review was conducted to explore recent advancements in the field of artificial intelligence (AI) in RCC pathology. The aim of this paper is to assess whether these advancements hold promise in improving the precision, efficiency, and objectivity of histopathological analysis for RCC, while also reducing costs and interobserver variability and potentially alleviating the labor and time burden experienced by pathologists. The reviewed AI-powered approaches demonstrate effective identification and classification abilities regarding several histopathological features associated with RCC, facilitating accurate diagnosis, grading, and prognosis prediction and enabling precise and reliable assessments. Nevertheless, implementing AI in renal cell carcinoma generates challenges concerning standardization, generalizability, benchmarking performance, and integration of data into clinical workflows. Developing methodologies that enable pathologists to interpret AI decisions accurately is imperative. Moreover, establishing more robust and standardized validation workflows is crucial to instill confidence in AI-powered systems' outcomes. These efforts are vital for advancing current state-of-the-art practices and enhancing patient care in the future.

The WIRE study a phase II, multi-arm, multi-centre, non-randomised window-of-opportunity clinical trial platform using a Bayesian adaptive design for proof-of-mechanism of novel treatment strategies in operable renal cell cancer - a study protocol

BACKGROUND

Window-of-opportunity trials, evaluating the engagement of drugs with their biological target in the time period between diagnosis and standard-of-care treatment, can help prioritise promising new systemic treatments for later-phase clinical trials. Renal cell carcinoma (RCC), the 7th commonest solid cancer in the UK, exhibits targets for multiple new systemic anti-cancer agents including DNA damage response inhibitors, agents targeting vascular pathways and immune checkpoint inhibitors. Here we present the trial protocol for the WIndow-of-opportunity clinical trial platform for evaluation of novel treatment strategies in REnal cell cancer (WIRE).

METHODS

WIRE is a Phase II, multi-arm, multi-centre, non-randomised, proof-of-mechanism (single and combination investigational medicinal product [IMP]), platform trial using a Bayesian adaptive design. The Bayesian adaptive design leverages outcome information from initial participants during pre-specified interim analyses to determine and minimise the number of participants required to demonstrate efficacy or futility. Patients with biopsy-proven, surgically resectable, cT1b+, cN0-1, cM0-1 clear cell RCC and no contraindications to the IMPs are eligible to participate. Participants undergo diagnostic staging CT and renal mass biopsy followed by treatment in one of the treatment arms for at least 14 days. Initially, the trial includes five treatment arms with cediranib, cediranib + olaparib, olaparib, durvalumab and durvalumab + olaparib. Participants undergo a multiparametric MRI before and after treatment. Vascularised and de-vascularised tissue is collected at surgery. A ≥ 30% increase in CD8+ T-cells on immunohistochemistry between the screening and nephrectomy is the primary endpoint for durvalumab-containing arms. Meanwhile, a reduction in tumour vascular permeability measured by Ktrans on dynamic contrast-enhanced MRI by ≥30% is the primary endpoint for other arms. Secondary outcomes include adverse events and tumour size change. Exploratory outcomes include biomarkers of drug mechanism and treatment effects in blood, urine, tissue and imaging.

DISCUSSION

WIRE is the first trial using a window-of-opportunity design to demonstrate pharmacological activity of novel single and combination treatments in RCC in the pre-surgical space. It will provide rationale for prioritising promising treatments for later phase trials and support the development of new biomarkers of treatment effect with its extensive translational agenda.

TRIAL REGISTRATION

ClinicalTrials.gov: NCT03741426 / EudraCT: 2018-003056-21 .

Integrative analysis of cross-modal features for the prognosis prediction of clear cell renal cell carcinoma

Motivation

As a highly heterogeneous disease, clear cell renal cell carcinoma (ccRCC) has quite variable clinical behaviors. The prognostic biomarkers play a crucial role in stratifying patients suffering from ccRCC to avoid over- and under-treatment. Researches based on hand-crafted features and single-modal data have been widely conducted to predict the prognosis of ccRCC. However, these experience-dependent methods, neglecting the synergy among multimodal data, have limited capacity to perform accurate prediction. Inspired by complementary information among multimodal data and the successful application of convolutional neural networks (CNNs) in medical image analysis, a novel framework was proposed to improve prediction performance.

Results

We proposed a cross-modal feature-based integrative framework, in which deep features extracted from computed tomography/histopathological images by using CNNs were combined with eigengenes generated from functional genomic data, to construct a prognostic model for ccRCC. Results showed that our proposed model can stratify high- and low-risk subgroups with significant difference (P-value < 0.05) and outperform the predictive performance of those models based on single-modality features in the independent testing cohort [C-index, 0.808 (0.728–0.888)]. In addition, we also explored the relationship between deep image features and eigengenes, and make an attempt to explain deep image features from the view of genomic data. Notably, the integrative framework is available to the task of prognosis prediction of other cancer with matched multimodal data.

Availability and implementation

https://github.com/zhang-de-lab/zhang-lab? from=singlemessage

Supplementary information

Supplementary data are available at Bioinformatics online.

Reproducibility and Crossplatform Validation of Reverse-Phase Protein Array Data

Reverse-phase protein array (RPPA) technology is a high-throughput antibody- and microarray-based approach for the rapid profiling of levels of proteins and protein posttranslational modifications in biological specimens. The technology consumes small amounts of samples, can sensitively detect low-abundance proteins and posttranslational modifications, enables measurements of multiple signaling pathways in parallel, has the capacity to analyze large sample numbers, and offers robust interexperimental reproducibility. These features of RPPA experiments have motivated and enabled the use of RPPA technology in various biomedical, translational, and clinical applications, including the delineation of molecular mechanisms of disease, profiling of druggable signaling pathway activation, and search for new prognostic markers. Owing to the complexity of many of these applications, such as developing multiplex protein assays for diagnostic laboratories or integrating posttranslational modification-level data using large-scale proteogenomic approaches, robust and well-validated data are essential. There are many distinct components of an RPPA workflow, and numerous possible technical setups and analysis parameter options exist. The differences between RPPA platform setups around the world offer opportunities to assess and minimize interplatform variation. Crossplatform validation may also aid in the evaluation of robust, platform-independent protein markers of disease and response to therapy.

Tumour heterogeneity in triplet-paired metastatic tumour tissues in metastatic renal cell carcinoma: concordance analysis of target gene sequencing data

AIMS

The aim of the present study was to determine the concordant correlation in the expression of 88 target genes from triple-paired metastatic tissues in individual patients with metastatic renal carcinoma (mRCC) using a target gene sequencing (TGS) approach.

METHODS

Between 2002 and 2017, a total of 350 triple-paired metastatic tissue samples from 262 patients with mRCC obtained from either nephrectomy or metastasectomy were used for TGS of 88 candidate genes. After quality check, 243 tissue samples from 81 patients were finally applied to TGS. The concordance of triple-paired tissues was analysed with the 88 TGS panels using bioinformatics tools.

RESULTS

Among 81 patients, alterations were observed in 42 (51.9%) for any of the 88 mRCC panel genes; however, no pathogenic gene was detected in 38 (39.5%) . Concordance >95% for altered gene expression among the three tissues was reported in 12 (28.6%) patients, while concordance >95% within two tissues was reported in 30 (71.4%); concordance <50% was reported in the remaining eight patients. Considering several types of genetic alterations, including deletions, insertions, missense and nonsense mutations, and splice variants, genes most frequently detected with genetic alterations in the patients with mRCC were PTEN loss, followed by FLCN, BCR, SMARCA2, AKAP9, MLH1, MYH11, APC and TP53.

CONCLUSIONS

The study provides reference information on the genetic alterations at various organ sites and the multi-heterogeneity of mRCC tissues. The concordance of pathogenic gene alterations within tissues was not high, and approximately half of the patients showed no pathogenic gene alterations at all.

Clinical utility of the S3-score for molecular prediction of outcome in non-metastatic and metastatic clear cell renal cell carcinoma

BACKGROUND

Stratification of cancer patients to identify those with worse prognosis is increasingly important. Through in silico analyses, we recently developed a gene expression-based prognostic score (S3-score) for clear cell renal cell carcinoma (ccRCC), using the cell type-specific expression of 97 genes within the human nephron. Herein, we verified the score using whole-transcriptome data of independent cohorts and extend its application for patients with metastatic disease receiving tyrosine kinase inhibitor treatment. Finally, we sought to improve the signature for clinical application using qRT-PCR.

METHODS

A 97 gene-based S3-score (S3₉₇) was evaluated in a set of 52 primary non-metastatic and metastatic ccRCC patients as well as in 53 primary metastatic tumors of sunitinib-treated patients. Gene expression data of The Cancer Genome Atlas (n = 463) was used for platform transfer and development of a simplified qRT-PCR-based 15-gene S3-score (S3₁₅). This S3₁₅-score was validated in 108 metastatic and non-metastatic ccRCC patients and ccRCC-derived metastases including in part several regions from one metastasis. Univariate and multivariate Cox regression stratified by T, N, M, and G were performed with cancer-specific and progression-free survival as primary endpoints.

RESULTS

The S3₉₇-score was significantly associated with cancer-specific survival (CSS) in 52 ccRCC patients (HR 2.9, 95% Cl 1.0-8.0, P_Log-rank = 3.3 × 10^-2) as well as progression-free survival in sunitinib-treated patients (2.1, 1.1-4.2, P_Log-rank = 2.2 × 10^-2). The qRT-PCR based S3₁₅-score performed similarly to the S3₉₇-score, and was significantly associated with CSS in our extended cohort of 108 patients (5.0, 2.1-11.7, P_Log-rank = 5.1 × 10^-5) including metastatic (9.3, 1.8-50.0, P_Log-rank = 2.3 × 10^-3) and non-metastatic patients (4.4, 1.2-16.3, P_Log-rank = 1.6 × 10^-2), even in multivariate Cox regression, including clinicopathological parameters (7.3, 2.5-21.5, P_Wald = 3.3 × 10^-4). Matched primary tumors and metastases revealed similar S3₁₅-scores, thus allowing prediction of outcome from metastatic tissue. The molecular-based qRT-PCR S3₁₅-score significantly improved prediction of CSS by the established clinicopathological-based SSIGN score (P = 1.6 × 10^-3).

CONCLUSION

The S3-score offers a new clinical avenue for ccRCC risk stratification in the non-metastatic, metastatic, and sunitinib-treated setting.