Corrie, P. | bioRxiv Science

bioRxiv2020

Tumour gene expression signature in primary melanoma predicts long-term outcomes: A prospective multicentre study

PurposePredicting outcomes after resection of primary melanoma remains crude, primarily based on tumour thickness. We explored gene expression signatures for their ability to better predict outcomes. MethodsDifferential expression analysis of 194 primary melanomas resected from patients who either developed distant metastasis (n=89) or did not (n=105) was performed. We identified 121 metastasis-associated genes that were included in our prognostic signature, "Cam_121". Several machine learning classification models were trained using nested leave- one-out cross validation (LOOCV) to test the signatures capacity to predict metastases, as well as regression models to predict survival. The prognostic accuracy was externally validated in two independent datasets. ResultsCam_121 performed significantly better in predicting distant metastases than any of the models trained with the clinical covariates alone (pAccuracy=4.92x10-3), as well as those trained with two published prognostic signatures. Cam_121 expression score was strongly associated with progression-free survival (HR=1.7, p=3.44x10-6), overall survival (HR=1.73, p=7.71x10-6) and melanoma-specific survival (HR=1.59, p=0.02). Cam_121 expression score also negatively correlated with measures of immune cell infiltration ({rho}=-0.73, p<2.2x10-16), with a higher score representing reduced tumour lymphocytic infiltration and a higher absolute 5-year risk of death in stage II melanoma. ConclusionsThe Cam_121 primary melanoma gene expression signature outperformed currently available alternatives in predicting the risk of distant recurrence. The signature confirmed (using unbiased approaches) the central prognostic importance of immune cell infiltration in long-term patient outcomes and could be used to identify stage II melanoma patients at highest risk of metastases and poor survival who might benefit most from adjuvant therapies. Translational relevancePredicting outcomes after resection of primary melanoma is currently based on traditional histopathological staging, however survival outcomes within these disease stages varies markedly. Since adjuvant systemic therapies are now being used routinely, accurate prognostic information is needed to better risk stratify patients and avoid unnecessary use of high cost, potentially harmful drugs, as well as to inform future adjuvant strategies. The Cam_121 gene expression signature appears to have this capability and warrants evaluation in prospective clinical trials.

bioinformatics↗

bioRxiv2020

Hotspot KRAS mutations in brain metastases at the first metastatic recurrence of cutaneous melanoma

IMPORTANCEBrain metastases occur in 60% of patients with advanced melanoma and are a major cause of melanoma-related mortality and morbidity. Although our understanding of the molecular alterations associated with melanoma progression is improving, there are currently no validated biomarkers which might help identify those patients at highest risk of developing brain metastases. OBJECTIVETo examine the somatic mutational and copy-number landscape of brain metastases that develop as the isolated first visceral site of recurrence - "early brain-metastasis" compared to extracranial melanoma metastases. DESIGN, SETTING AND PARTICIPANTSWhole-exome sequencing of 50 tumors from patients undergoing surgical resection of one or more brain metastasis occurring as the first site of visceral relapse were identified from prospectively maintained databases in Sydney, Wellington, New York and Cambridge. Whole exome sequencing analyses allowed mutational profiles to be compared to cutaneous melanomas in The Cancer Genome Atlas (SKCM-TCGA; n=358) and the Memorial Sloan Kettering (SKCM-MSK-IMPACT; n=186) datasets. An external dataset comprising a further 18 patients with surgically resected early brain metastasis from two additional academic centers served as an independent validation cohort. MAIN OUTCOMES AND MEASURESTo assess the frequency of driver mutations in early brain metastasis and their influence on survival. RESULTSIn concordance with the landmark melanoma sequencing studies, we identified mutations in BRAF (21/50, 42%), NRAS (14/50, 28%) and NF1 (11/50, 22%) as the most frequently mutated melanoma driver genes. When compared to the mutational landscape of cutaneous melanomas in TCGA (SKCM-TCGA), KRAS was the most significantly enriched driver gene, with 5/50 (10%) of brain metastases harboring non-synonymous mutations, of which 4/5 (80%) were in the hotspot positions of codons 12 and 61. This was significantly higher than the corresponding frequency of KRAS-mutations within the entire SKCM-TCGA (2% (7/358), p=0.009, Fishers Exact Test) as well as the SKCM-MSK-IMPACT cohort (1.6% (3/186), p=0.016). Variants in KRAS were mutually exclusive from BRAFV600, NRAS and HRAS mutations and were associated with a significantly reduced overall survival from resection of brain metastasis (relative to KRAS-wild type brain metastases) in multivariate Cox proportional hazard models (HR 1.80, 95% CI 1.46-24.89, p=0.013). Mutations in KRAS were also clonal and concordant with extracranial disease, which suggests these mutations are present within the primary tumor CONCLUSIONS AND RELEVANCEOur analysis, the largest to date, suggests that early metastases to the brain (presenting as the first site of visceral relapse) are characterized by significant enrichment of hotspot KRAS mutations, potentially implicating constitutive RAS-driven cellular programs in neurotropic metastatic behavior in these cases. Based on these data, we suggest that screening for KRAS mutations might help identify those patients with primary melanoma at higher risk of brain metastases or poor survival, and could help inform future surveillance strategies. Key PointsO_ST_ABSQuestionC_ST_ABSWhat is the frequency of driver mutations in early melanoma brain metastases? FindingsIn this study of 50 patients with melanoma metastasizing first to the brain, KRAS mutations were the most significantly enriched driver gene (n=5, 10% of patients) when compared to landmark cutaneous melanoma studies. The high KRAS mutation frequency was also observed in an external validation cohort of 18 patients with early brain metastases. Mutations in KRAS were mutually exclusive from mutations in the key RAS signaling genes and conferred a worse overall survival from resection of brain metastasis. MeaningHotspot KRAS mutations could help identify those patients with primary melanoma at higher risk of brain metastases that may benefit from more intensive, protracted surveillance as well as earlier use of adjuvant therapy.

genomics↗

bioRxiv2019

Multi-site clonality analyses uncovers pervasive subclonal heterogeneity and branching evolution across melanoma metastases

Metastatic melanoma carries a poor prognosis despite modern systemic therapies. Understanding the evolution of the disease could help inform patient management. Through whole-genome sequencing of 13 melanoma metastases sampled at autopsy from a treatment naive patient and by leveraging the analytical power of multi-sample analyses, we reveal that metastatic cells may depart the primary tumour very early in the disease course and follow a branched pattern of evolution. Truncal UV-induced mutations that often swamp downstream analyses of heterogeneity, were found to be replaced by APOBEC-associated mutations in the branches of the evolutionary tree. Multi-sample analyses from a further 7 patients confirmed that branched evolution was pervasive, representing an important mode of melanoma dissemination. Our analyses illustrate that combining cancer cell fraction estimates across multiple metastases provides higher resolution phylogenetic reconstructions relative to single sample analyses and highlights the limitations of accurately inferring inter-tumoural heterogeneity from a single biopsy.

genomics↗