Background: Research grade Fresh Frozen (FF) DNA material is not yet routinely collected in clinical practice. Many hospitals, however, collect and store Formalin Fixed Paraffin Embedded (FFPE) tumor samples. Consequently, the sample size of whole genome cancer cohort studies could be increased tremendously by including FFPE samples, although the presence of artefacts might obfuscate the variant calling. To assess whether FFPE material can be used for cohort studies, we performed an in-depth comparison of somatic SNVs called on matching FF and FFPE Whole Genome Sequence (WGS) samples extracted from the same tumor. Methods: Four variant callers (i.e. Strelka2, Mutect2, VarScan2 and Shimmer) were used to call somatic variants on matching FF and FFPE WGS samples from a metastatic prostate tumor. Using the variants identified by these callers, we developed a heuristic to maximize the overlap between the FF and its FFPE counterpart in terms of sensitivity and precision. The proposed variant calling approach was then validated on nine matched primary samples. Finally, we assessed what fraction of the discrepancy could be attributed to intra-tumor heterogeneity (ITH), by comparing the overlap in clonal and subclonal somatic variants. Results: We first compared variants between an FF and an FFPE sample from a metastatic prostate tumor, showing that on average 50% of the calls in the FF are recovered in the FFPE sample, with notable differences between callers. Combining the variants of the different callers using a simple heuristic, increases both the precision and the sensitivity of the variant calling. Validating the heuristic on nine additional matched FF-FFPE samples, resulted in an average F1-score of 0.58 and an outperformance of any of the individual callers. In addition, we could show that part of the discrepancy between the FF and the FFPE samples can be attributed to ITH. Conclusion: This study illustrates that when using the correct variant calling strategy, the majority of clonal SNVs can be recovered in an FFPE sample with high precision and sensitivity. These results suggest that somatic variants derived from WGS of FFPE material can be used in cohort studies.

In prostate cancer, there is an urgent need for objective prognostic biomarkers that identify the metastatic potential of a tumor at an early stage. While recent analyses indicated TP53 mutations as candidate biomarkers, molecular profiling in a clinical setting is complicated by tumor heterogeneity. Deep learning models that predict the spatial presence of TP53 mutations in whole slide images (WSI) offer the potential to mitigate this issue. To assess the potential of WSIs as proxies for spatially resolved profiling and as biomarkers for aggressive disease, we developed TiDo, a deep learning model that achieves state-of-the-art performance in predicting TP53 mutations from WSIs of primary prostate tumors. In an independent multifocal cohort, the model showed successful generalization at both the patient and lesion level. Analysis of model predictions revealed that false positive (FP) predictions could at least partially be explained by TP53 deletions, suggesting that some FP carry an alteration that leads to the same histological phenotype as TP53 mutations. Comparative expression and histologic cell type analyses identified a TP53-like cellular phenotype triggered by expression of pathways affecting stromal composition. Together, these findings indicate that WSI-based models might not be able to perfectly predict the spatial presence of individual TP53 mutations but they have the potential to elucidate the prognosis of a tumor by depicting a downstream phenotype associated with aggressive disease biomarkers. Significance: Deep learning models predicting TP53 mutations from whole slide images of prostate cancer capture histologic phenotypes associated with stromal composition, lymph node metastasis, and biochemical recurrence, indicating their potential as in silico prognostic biomarkers. See related commentary by Bordeleau, p. 2809

Most known driver genes of metastatic prostate cancer are frequently mutated. To dig into the long tail of rarely mutated drivers, we performed network-based driver identification on the Hartwig Medical Foundation metastatic prostate cancer data set (HMF cohort). Hereto, we developed GoNetic, a method based on probabilistic pathfinding, to identify recurrently mutated subnetworks. In contrast to most state-of-the-art network-based methods, GoNetic can leverage sample-specific mutational information and the weights of the underlying prior network. When applied to the HMF cohort, GoNetic successfully recovered known primary and metastatic drivers of prostate cancer that are frequently mutated in the HMF cohort (TP53, RB1, and CTNNB1). In addition, the identified subnetworks contain frequently mutated genes, reflect processes related to metastatic prostate cancer, and contain rarely mutated driver candidates. To further validate these rarely mutated genes, we assessed whether the identified genes were more mutated in metastatic than in primary samples using an independent cohort. Then we evaluated their association with tumor evolution and with the lymph node status of the patients. This resulted in forwarding several novel putative driver genes for metastatic prostate cancer, some of which might be prognostic for disease evolution.

We present deep link prediction (DLP), a method for the interpretation of loss-of-function screens. Our approach uses representation-based link prediction to reprioritize phenotypic readouts by integrating screening experiments with gene-gene interaction networks. We validate on 2 different loss-of-function technologies, RNAi and CRISPR, using datasets obtained from DepMap. Extensive benchmarking shows that DLP-DeepWalk outperforms other methods in recovering cell-specific dependencies, achieving an average precision well above 90% across 7 different cancer types and on both RNAi and CRISPR data. We show that the genes ranked highest by DLP-DeepWalk are appreciably more enriched in drug targets compared to the ranking based on original screening scores. Interestingly, this enrichment is more pronounced on RNAi data compared to CRISPR data, consistent with the greater inherent noise of RNAi screens. Finally, we demonstrate how DLP-DeepWalk can infer the molecular mechanism through which putative targets trigger cell line mortality.

In prostate cancer, there is an urgent need for objective prognostic biomarkers that identify the metastatic potential of a tumor at an early stage. While recent analyses indicated TP53 mutations as candidate biomarkers, molecular profiling in a clinical setting is complicated by tumor heterogeneity. Deep learning models that predict the spatial presence of TP53 mutations in whole slide images (WSI) offer the potential to mitigate this issue. To assess the potential of WSIs as proxies for spatially resolved profiling and as biomarkers for aggressive disease, we developed TiDo, a deep learning model that achieves state-of-the-art performance in predicting TP53 mutations from WSIs of primary prostate tumors. In an independent multifocal cohort, the model showed successful generalization at both the patient and lesion level. Analysis of model predictions revealed that false positive (FP) predictions could at least partially be explained by TP53 deletions, suggesting that some FP carry an alteration that leads to the same histological phenotype as TP53 mutations. Comparative expression and histologic cell type analyses identified a TP53-like cellular phenotype triggered by expression of pathways affecting stromal composition. Together, these findings indicate that WSI-based models might not be able to perfectly predict the spatial presence of individual TP53 mutations but they have the potential to elucidate the prognosis of a tumor by depicting a downstream phenotype associated with aggressive disease biomarkers.Significance:Deep learning models predicting TP53 mutations from whole slide images of prostate cancer capture histologic phenotypes associated with stromal composition, lymph node metastasis, and biochemical recurrence, indicating their potential as in silico prognostic biomarkers.See related commentary by Bordeleau, p. 2809

Background. Research grade Fresh Frozen (FF) DNA material is not yet routinely collected in clinical practice. Many hospitals, however, collect and store Formalin Fixed Paraffin Embedded (FFPE) tumor samples. Consequently, the sample size of whole genome cancer cohort studies could be increased tremendously by including FFPE samples, although the presence of artefacts might obfuscate the variant calling. To assess whether FFPE material can be used for cohort studies, we performed an in-depth comparison of somatic SNVs called on matching FF and FFPE Whole Genome Sequence (WGS) samples extracted from the same tumor. Results. We first compared the calls between an FF and an FFPE from a metastatic prostate tumor, showing that on average 50% of the calls in the FF are recovered in the FFPE sample, with notable differences between variant callers. Combining the variants of the different callers using a simple heuristic increases both the precision and the sensitivity of the variant calling. Validating the heuristic on nine additional matched FF-FFPE samples, resulted in an average F1-score of 0.58 and an outperformance of any of the individual callers. In addition, we could show that part of the discrepancy between the FF and the FFPE samples can be attributed to intra-tumor heterogeneity (ITH). Conclusion. This study illustrates that when using the correct variant calling strategy, the majority of clonal SNVs can be recovered in an FFPE sample with high precision and sensitivity. These results suggest that somatic variants derived from WGS of FFPE material can be used in cohort studies.