Wang, Xuefeng

ImportanceWhile the incidence of early-onset metastatic colorectal cancer (mCRC) has been increasing, studies on the age-related disparity in this group of patients are limited.ObjectiveTo evaluate the association of age with treatment-related adverse events and survival in patients with mCRC and explore the potential underlying factors.Design, Setting, and ParticipantsThis cohort study included 1959 individuals. Individual data on 1223 patients with mCRC who received first-line fluorouracil and oxaliplatin therapy in 3 clinical trials, and clinical and genomic data of 736 patients with mCRC from Moffitt Cancer Center were used to assess genomic alterations and serve as an external validation cohort. All statistical analyses were conducted from October 1, 2021, through November 12, 2022.ExposuresMetastatic colorectal cancer.Main Outcomes and MeasuresSurvival outcomes and treatment-related adverse events were compared among patients in 3 age groups: younger than 50 (early onset), 50 to 65, and older than 65 years.ResultsIn the total population of 1959 individuals, 1145 (58.4%) were men. Among 1223 patients from previous clinical trials, 179 (14.6%) in the younger than 50 years group, 582 (47.6%) in the 50 to 65 years group, and 462 (37.8%) in the older than 65 years group had similar baseline characteristics except for sex and race. The younger than 50 years group had significantly shorter progression-free survival (PFS) (hazard ratio [HR], 1.46; 95% CI, 1.22-1.76; P < .001) and overall survival (OS) (HR, 1.48; 95% CI, 1.19-1.84; P < .001) compared with the 50 to 65 years group after adjustment for sex, race, and performance status. Significantly shorter OS in the younger than 50 years group was confirmed in the Moffitt cohort. The younger than 50 years group had a significantly higher incidence of nausea and vomiting (69.3% vs 57.6% [50-65 years] vs 60.4% [>65 years]; P = .02), severe abdominal pain (8.4% vs 3.4% vs 3.5%; P = .02), severe anemia (6.1% vs 1.0% vs 1.5%; P < .001), and severe rash (2.8% vs 1.2% vs 0.4% P = .047). The younger than 50 years group also had earlier onset of nausea and vomiting (1.0 vs 2.1 vs 2.6 weeks; P = .01), mucositis (3.6 vs 5.1 vs 5.7 weeks; P = .05), and neutropenia (8.0 vs 9.4 vs 8.4 weeks; P = .04), and shorter duration of mucositis (0.6 vs 0.9 vs 1.0 weeks; P = .006). In the younger than 50 years group, severe abdominal pain and severe liver toxic effects were associated with shorter survival. The Moffitt genomic data showed that the younger than 50 years group had a higher prevalence of CTNNB1 mutation (6.6% vs 3.1% vs 2.3%; P = .047), ERBB2 amplification (5.1% vs 0.6% vs 2.3%; P = .005), and CREBBP mutation (3.1% vs 0.9% vs 0.5%; P = .05), but lower prevalence of BRAF mutation (7.7% vs 8.5% vs 16.7%; P = .002).Conclusions and RelevanceIn this cohort study of 1959 patients, those with early-onset mCRC showed worse survival outcomes and unique adverse event patterns, which could be partially attributed to distinct genomic profiles. These findings may inform individualized management approaches in patients with early-onset mCRC.

Normalization is the first critical step in microbiome sequencing data analysis used to account for variable library sizes. Current RNA-Seq based normalization methods that have been adapted for microbiome data fail to consider the unique characteristics of microbiome data, which contain a vast number of zeros due to the physical absence or under-sampling of the microbes. Normalization methods that specifically address the zero-inflation remain largely undeveloped. Here we propose geometric mean of pairwise ratios—a simple but effective normalization method—for zero-inflated sequencing data such as microbiome data. Simulation studies and real datasets analyses demonstrate that the proposed method is more robust than competing methods, leading to more powerful detection of differentially abundant taxa and higher reproducibility of the relative abundances of taxa.

Normalization is the first critical step in microbiome sequencing data analysis used to account for variable library sizes. Current RNA-Seq based normalization methods that have been adapted for microbiome data fail to consider the unique characteristics of microbiome data, which contain a vast number of zeros due to the physical absence or under-sampling of the microbes. Normalization methods that specifically address the zero inflation remain largely undeveloped. Here we propose GMPR - a simple but effective normalization method - for zero-inflated sequencing data such as microbiome data. Simulation studies and real datasets analyses demonstrate that the proposed method is more robust than competing methods, leading to more powerful detection of differentially abundant taxa and higher reproducibility of the relative abundances of taxa.

Classic multinomial logit model, commonly used in multiclass regression problem, is restricted to few predictors and does not take into account the relationship among variables. It has limited use for genomic data, where the number of genomic features far exceeds the sample size. Genomic features such as gene expressions are usually related by an underlying biological network. Efficient use of the network information is important to improve classification performance as well as the biological interpretability. We proposed a multinomial logit model that is capable of addressing both the high dimensionality of predictors and the underlying network information. Group lasso was used to induce model sparsity, and a network-constraint was imposed to induce the smoothness of the coefficients with respect to the underlying network structure. To deal with the non-smoothness of the objective function in optimization, we developed a proximal gradient algorithm for efficient computation. The proposed model was compared to models with no prior structure information in both simulations and a problem of cancer subtype prediction with real TCGA (the cancer genome atlas) gene expression data. The network-constrained mode outperformed the traditional ones in both cases.