Q-omics provides the consensus-scored GPT2 profile across patient tissues and cancer cell-line models. GPT2 expression is associated with patient survival in 22 of 34 cancer types, with the highest sampling consensus in SCLC. Among the 18 cancer types available for tumor–normal comparison, GPT2 is differentially expressed in 15, with the highest sampling consensus in HNSC. Additionally, GPT2 RNA expression shows 20,239 significant protein co-abundance associations, with the highest sampling consensus in LSCC. Together, these results highlight SCLC, HNSC, and LSCC as cancer lineages where GPT2 shows reproducible signals across survival, tumor–normal expression, and patient cross-omics analyses.
Every result is evaluated using two consensus scores. Sampling consensus measures how consistently a finding is reproduced within a cancer lineage across different conditions. Lineage consensus measures how broadly the result is shared across cancer types, distinguishing pan-cancer signals from lineage-specific patterns.
Premium analyses for GPT2 — synthetic lethality, tumor antigen, and pembrolizumab response.
This table summarizes GPT2 survival associations across molecular data types. GPT2 RNA expression shows survival associations in the most cancer types (22), followed by mutation status (5) and mass-spec protein abundance (5). The rightmost column indicates the cancer type with the highest sampling consensus for each molecular layer.
This table ranks reproducible GPT2 RNA expression–survival associations across cancer types. High GPT2 expression shows unfavorable associations in SARC, ACC and COAD, but favorable associations in SCLC, OV and LIHC. The SCLC Kaplan–Meier curve shows clear separation, with the low-expression group declining faster, consistent with the favorable association (log-rank p < 0.001). Together, the overview and detailed table identify SCLC as the clearest survival context for GPT2 RNA expression.
This table summarizes GPT2 tumor–normal expression differences by data type. RNA shows broader differences across cancer types, with a lineage consensus of 15, while mass-spec protein shows differences in 6. The strongest signals are observed in HNSC for RNA and HNSC for protein.
This table ranks reproducible tumor–normal expression differences for GPT2. A negative fold-change indicates higher expression in normal tissue than in tumor tissue. GPT2 shows lower tumor expression in HNSC, THCA and KIRC and higher tumor expression in LUAD, COAD and LUSC. The HNSC box plot shows higher GPT2 RNA expression in normal versus tumor tissue (log2 FC = −1.637, t-test p < 0.001).
This table shows molecular features associated with GPT2 in patient tissues and cancer cell lines. In patient samples, GPT2 shows the broadest associations at the RNA and protein expression levels, with LSCC recurring as the lineage with the largest associated feature set. In cancer cell lines, GPT2 RNA and mutation anchors are most strongly linked to RNA-expression features, especially in URINARY_TRACT, while CRISPR and shRNA rows add functional-dependency signals in UPPER_AERODIGESTIVE_TRACT and SOFT_TISSUE.
