Noninvasive Early Identification of Therapeutic Benefit from Immune Checkpoint Inhibition.

Although treatment of non-small cell lung cancer (NSCLC) with immune checkpoint inhibitors (ICIs) can produce remarkably durable responses, most patients develop early disease progression. Furthermore, initial response assessment by conventional imaging is often unable to identify which patients will achieve durable clinical benefit (DCB). Here, we demonstrate that pre-treatment circulating tumor DNA (ctDNA) and peripheral CD8 T cell levels are independently associated with DCB. We further show that ctDNA dynamics after a single infusion can aid in identification of patients who will achieve DCB. Integrating these determinants, we developed and validated an entirely noninvasive multiparameter assay (DIREct-On, Durable Immunotherapy Response Estimation by immune profiling and ctDNA-On-treatment) that robustly predicts which patients will achieve DCB with higher accuracy than any individual feature. Taken together, these results demonstrate that integrated ctDNA and circulating immune cell profiling can provide accurate, noninvasive, and early forecasting of ultimate outcomes for NSCLC patients receiving ICIs.

CD4 T cells and toxicity from immune checkpoint blockade.

Immune-related toxicities, otherwise known as immune-related adverse events (irAEs), occur in a substantial fraction of cancer patients treated with immune checkpoint inhibitors (ICIs). Ranging from asymptomatic to life-threatening, ICI-induced irAEs can result in hospital admission, high-dose corticosteroid treatment, ICI discontinuation, and in some cases, death. A deeper understanding of the factors underpinning severe irAE development will be essential for improved irAE prediction and prevention, toward maximizing the benefits and safety profiles of ICIs. In recent work, we applied mass cytometry, single-cell RNA sequencing, single-cell V(D)J sequencing, bulk RNA sequencing, and bulk T-cell receptor (TCR) sequencing to identify pretreatment determinants of severe irAE development in patients with advanced melanoma. Across 71 patients separated into three cohorts, we found that two baseline features in circulation-elevated activated CD4 effector memory T-cell abundance and TCR diversity-are associated with severe irAE development, independent of the affected organ system within 3 months of ICI treatment initiation. Here, we provide an extended perspective on this work, synthesize and discuss related literature, and summarize practical considerations for clinical translation. Collectively, these findings lay a foundation for data-driven and mechanistic insights into irAE development, with the potential to reduce ICI morbidity and mortality in the future.

Integrating genomic features for non-invasive early lung cancer detection.

Radiologic screening of high-risk adults reduces lung-cancer-related mortality1,2; however, a small minority of eligible individuals undergo such screening in the United States3,4. The availability of blood-based tests could increase screening uptake. Here we introduce improvements to cancer personalized profiling by deep sequencing (CAPP-Seq)5, a method for the analysis of circulating tumour DNA (ctDNA), to better facilitate screening applications. We show that, although levels are very low in early-stage lung cancers, ctDNA is present prior to treatment in most patients and its presence is strongly prognostic. We also find that the majority of somatic mutations in the cell-free DNA (cfDNA) of patients with lung cancer and of risk-matched controls reflect clonal haematopoiesis and are non-recurrent. Compared with tumour-derived mutations, clonal haematopoiesis mutations occur on longer cfDNA fragments and lack mutational signatures that are associated with tobacco smoking. Integrating these findings with other molecular features, we develop and prospectively validate a machine-learning method termed 'lung cancer likelihood in plasma' (Lung-CLiP), which can robustly discriminate early-stage lung cancer patients from risk-matched controls. This approach achieves performance similar to that of tumour-informed ctDNA detection and enables tuning of assay specificity in order to facilitate distinct clinical applications. Our findings establish the potential of cfDNA for lung cancer screening and highlight the importance of risk-matching cases and controls in cfDNA-based screening studies.

Healthcare disparities, screening, and molecular testing in the changing landscape of non-small cell lung cancer in the United States: a review.

Inequitable access to care continues to hinder improvements in diagnosis and treatment of lung cancer. This review describes healthcare disparities in the changing landscape of non-small cell lung cancer (NSCLC) in the United States, focusing on racial, ethnic, sex-based, and socioeconomic trends. Furthermore, strategies to address disparities, overcome challenges, and improve patient outcomes are proposed. Barriers exist across lung cancer screening, diagnosis, and treatment regimens, varying by sex, age, race and ethnicity, geography, and socioeconomic status. Incidence and mortality rates of lung cancer are higher among Black men than White men, and incidences in young women are substantially greater than in young men. Disparities may be attributed to geographic differences in screening access, with correlating higher incidence and mortality rates in rural versus urban areas. Lower socioeconomic status is also linked to lower survival rates. Several strategies could help reduce disparities and improve outcomes. Current guidelines could improve screening eligibility by incorporating sex, race, and socioeconomic status variables. Patient and clinician education on screening guidelines and patient-level barriers to care are key, and biomarker testing is critical since ~ 70% of patients with NSCLC have an actionable biomarker. Timely diagnosis, staging, and comprehensive biomarker testing, including cell-free DNA liquid biopsy, may provide valuable treatment guidance for patients with NSCLC. Efforts to improve lung cancer screening and biomarker testing access, decrease bias, and improve education about screening and testing are needed to reduce healthcare disparities in NSCLC.

PACT: a pipeline for analysis of circulating tumor DNA.

MOTIVATION: Detection of genomic alterations in circulating tumor DNA (ctDNA) is currently used for active clinical monitoring of cancer progression and treatment response. While methods for analysis of small mutations are more developed, strategies for detecting structural variants (SVs) in ctDNA are limited. Additionally, reproducibly calling small-scale mutations, copy number alterations, and SVs in ctDNA is challenging due to the lack to unified tools for these different classes of variants. RESULTS: We developed a unified pipeline for the analysis of ctDNA [Pipeline for the Analysis of ctDNA (PACT)] that accurately detects SVs and consistently outperformed similar tools when applied to simulated, cell line, and clinical data. We provide PACT in the form of a Common Workflow Language pipeline which can be run by popular workflow management systems in high-performance computing environments. AVAILABILITY AND IMPLEMENTATION: PACT is freely available at https://github.com/ChrisMaherLab/PACT.

Tumorigenic circulating tumor cells from xenograft mouse models of non-metastatic NSCLC patients reveal distinct single cell heterogeneity and drug responses.

BACKGROUND: Circulating tumor cells (CTCs) are liquid biopsies that represent micrometastatic disease and may offer unique insights into future recurrences in non-small cell lung cancer (NSCLC). Due to CTC rarity and limited stability, no stable CTC-derived xenograft (CDX) models have ever been generated from non-metastatic NSCLC patients directly. Alternative strategies are needed to molecularly characterize CTCs and means of potential future metastases in this potentially curable patient group. METHODS: Surgically resected NSCLC primary tumor tissues from non-metastatic patients were implanted subcutaneously in immunodeficient mice to establish primary tumor patient-derived xenograft (ptPDX) models. CTCs were isolated as liquid biopsies from the blood of ptPDX mice and re-implanted subcutaneously into naïve immunodeficient mice to generate liquid biopsy CTC-derived xenograft (CDX) tumor models. Single cell RNA sequencing was performed and validated in an external dataset of non-xenografted human NSCLC primary tumor and metastases tissues. Drug response testing in CDX models was performed with standard of care chemotherapy (carboplatin/paclitaxel). Blockade of MYC, which has a known role in drug resistance, was performed with a MYC/MAX dimerization inhibitor (10058-F4). RESULTS: Out of ten ptPDX, two (20%) stable liquid biopsy CDX mouse models were generated. Single cell RNA sequencing analysis revealed an additional regenerative alveolar epithelial type II (AT2)-like cell population in CDX tumors that was also identified in non-xenografted NSCLC patients' metastases tissues. Drug testing using these CDX models revealed different treatment responses to carboplatin/paclitaxel. MYC target genes and c-MYC protein were upregulated in the chemoresistant CDX model, while MYC/MAX dimerization blocking could overcome chemoresistance to carboplatin/paclitaxel. CONCLUSIONS: To overcome the lack of liquid biopsy CDX models from non-metastatic NSCLC patients, CDX models can be generated with CTCs from ptPDX models that were originally established from patients' primary tumors. Single cell analyses can identify distinct drug responses and cell heterogeneities in CDX tumors that can be validated in NSCLC metastases tissues. CDX models deserve further development and study to discover personalized strategies against micrometastases in non-metastatic NSCLC patients.

Single Circulating-Tumor-Cell-Targeted Sequencing to Identify Somatic Variants in Liquid Biopsies in Non-Small-Cell Lung Cancer Patients.

Non-small-cell lung cancer (NSCLC) accounts for most cancer-related deaths worldwide. Liquid biopsy by a blood draw to detect circulating tumor cells (CTCs) is a tool for molecular profiling of cancer using single-cell and next-generation sequencing (NGS) technologies. The aim of the study was to identify somatic variants in single CTCs isolated from NSCLC patients by targeted NGS. Thirty-one subjects (20 NSCLC patients, 11 smokers without cancer) were enrolled for blood draws (7.5 mL). CTCs were identified by immunofluorescence, individually retrieved, and DNA-extracted. Targeted NGS was performed to detect somatic variants (single-nucleotide variants (SNVs) and insertions/deletions (Indels)) across 65 oncogenes and tumor suppressor genes. Cancer-associated variants were classified using OncoKB database. NSCLC patients had significantly higher CTC counts than control smokers (p = 0.0132; Mann-Whitney test). Analyzing 23 CTCs and 13 white blood cells across seven patients revealed a total of 644 somatic variants that occurred in all CTCs within the same subject, ranging from 1 to 137 per patient. The highest number of variants detected in ≥1 CTC within a patient was 441. A total of 18/65 (27.7%) genes were highly mutated. Mutations with oncogenic impact were identified in functional domains of seven oncogenes/tumor suppressor genes (NF1, PTCH1, TP53, SMARCB1, SMAD4, KRAS, and ERBB2). Single CTC-targeted NGS detects heterogeneous and shared mutational signatures within and between NSCLC patients. CTC single-cell genomics have potential for integration in NSCLC precision oncology.

Cell-free DNA ultra-low-pass whole genome sequencing to distinguish malignant peripheral nerve sheath tumor (MPNST) from its benign precursor lesion: A cross-sectional study.

BACKGROUND: The leading cause of mortality for patients with the neurofibromatosis type 1 (NF1) cancer predisposition syndrome is the development of malignant peripheral nerve sheath tumor (MPNST), an aggressive soft tissue sarcoma. In the setting of NF1, this cancer type frequently arises from within its common and benign precursor, plexiform neurofibroma (PN). Transformation from PN to MPNST is challenging to diagnose due to difficulties in distinguishing cross-sectional imaging results and intralesional heterogeneity resulting in biopsy sampling errors. METHODS AND FINDINGS: This multi-institutional study from the National Cancer Institute and Washington University in St. Louis used fragment size analysis and ultra-low-pass whole genome sequencing (ULP-WGS) of plasma cell-free DNA (cfDNA) to distinguish between MPNST and PN in patients with NF1. Following in silico enrichment for short cfDNA fragments and copy number analysis to estimate the fraction of plasma cfDNA originating from tumor (tumor fraction), we developed a noninvasive classifier that differentiates MPNST from PN with 86% pretreatment accuracy (91% specificity, 75% sensitivity) and 89% accuracy on serial analysis (91% specificity, 83% sensitivity). Healthy controls without NF1 (participants = 16, plasma samples = 16), PN (participants = 23, plasma samples = 23), and MPNST (participants = 14, plasma samples = 46) cohorts showed significant differences in tumor fraction in plasma (P = 0.001) as well as cfDNA fragment length (P < 0.001) with MPNST samples harboring shorter fragments and being enriched for tumor-derived cfDNA relative to PN and healthy controls. No other covariates were significant on multivariate logistic regression. Mutational analysis demonstrated focal NF1 copy number loss in PN and MPNST patient plasma but not in healthy controls. Greater genomic instability including alterations associated with malignant transformation (focal copy number gains in chromosome arms 1q, 7p, 8q, 9q, and 17q; focal copy number losses in SUZ12, SMARCA2, CDKN2A/B, and chromosome arms 6p and 9p) was more prominently observed in MPNST plasma. Furthermore, the sum of longest tumor diameters (SLD) visualized by cross-sectional imaging correlated significantly with paired tumor fractions in plasma from MPNST patients (r = 0.39, P = 0.024). On serial analysis, tumor fraction levels in plasma dynamically correlated with treatment response to therapy and minimal residual disease (MRD) detection before relapse. Study limitations include a modest MPNST sample size despite accrual from 2 major referral centers for this rare malignancy, and lack of uniform treatment and imaging protocols representing a real-world cohort. CONCLUSIONS: Tumor fraction levels derived from cfDNA fragment size and copy number alteration analysis of plasma cfDNA using ULP-WGS significantly correlated with MPNST tumor burden, accurately distinguished MPNST from its benign PN precursor, and dynamically correlated with treatment response. In the future, our findings could form the basis for improved early cancer detection and monitoring in high-risk cancer-predisposed populations.

Urine tumor DNA detection of minimal residual disease in muscle-invasive bladder cancer treated with curative-intent radical cystectomy: A cohort study.

BACKGROUND: The standard of care treatment for muscle-invasive bladder cancer (MIBC) is radical cystectomy, which is typically preceded by neoadjuvant chemotherapy. However, the inability to assess minimal residual disease (MRD) noninvasively limits our ability to offer bladder-sparing treatment. Here, we sought to develop a liquid biopsy solution via urine tumor DNA (utDNA) analysis. METHODS AND FINDINGS: We applied urine Cancer Personalized Profiling by Deep Sequencing (uCAPP-Seq), a targeted next-generation sequencing (NGS) method for detecting utDNA, to urine cell-free DNA (cfDNA) samples acquired between April 2019 and November 2020 on the day of curative-intent radical cystectomy from 42 patients with localized bladder cancer. The average age of patients was 69 years (range: 50 to 86), of whom 76% (32/42) were male, 64% (27/42) were smokers, and 76% (32/42) had a confirmed diagnosis of MIBC. Among MIBC patients, 59% (19/32) received neoadjuvant chemotherapy. utDNA variant calling was performed noninvasively without prior sequencing of tumor tissue. The overall utDNA level for each patient was represented by the non-silent mutation with the highest variant allele fraction after removing germline variants. Urine was similarly analyzed from 15 healthy adults. utDNA analysis revealed a median utDNA level of 0% in healthy adults and 2.4% in bladder cancer patients. When patients were classified as those who had residual disease detected in their surgical sample (n = 16) compared to those who achieved a pathologic complete response (pCR; n = 26), median utDNA levels were 4.3% vs. 0%, respectively (p = 0.002). Using an optimal utDNA threshold to define MRD detection, positive utDNA MRD detection was highly correlated with the absence of pCR (p < 0.001) with a sensitivity of 81% and specificity of 81%. Leave-one-out cross-validation applied to the prediction of pathologic response based on utDNA MRD detection in our cohort yielded a highly significant accuracy of 81% (p = 0.007). Moreover, utDNA MRD-positive patients exhibited significantly worse progression-free survival (PFS; HR = 7.4; 95% CI: 1.4-38.9; p = 0.02) compared to utDNA MRD-negative patients. Concordance between urine- and tumor-derived mutations, determined in 5 MIBC patients, was 85%. Tumor mutational burden (TMB) in utDNA MRD-positive patients was inferred from the number of non-silent mutations detected in urine cfDNA by applying a linear relationship derived from The Cancer Genome Atlas (TCGA) whole exome sequencing of 409 MIBC tumors. We suggest that about 58% of these patients with high inferred TMB might have been candidates for treatment with early immune checkpoint blockade. Study limitations included an analysis restricted only to single-nucleotide variants (SNVs), survival differences diminished by surgery, and a low number of DNA damage response (DRR) mutations detected after neoadjuvant chemotherapy at the MRD time point. CONCLUSIONS: utDNA MRD detection prior to curative-intent radical cystectomy for bladder cancer correlated significantly with pathologic response, which may help select patients for bladder-sparing treatment. utDNA MRD detection also correlated significantly with PFS. Furthermore, utDNA can be used to noninvasively infer TMB, which could facilitate personalized immunotherapy for bladder cancer in the future.

Correction: Urine tumor DNA detection of minimal residual disease in muscle-invasive bladder cancer treated with curative-intent radical cystectomy: A cohort study.

[This corrects the article DOI: 10.1371/journal.pmed.1003732.].

Circulating Tumor DNA Analysis for Detection of Minimal Residual Disease After Chemoradiotherapy for Localized Esophageal Cancer.

BACKGROUND & AIMS: Biomarkers are needed to risk stratify after chemoradiotherapy for localized esophageal cancer. These could improve identification of patients at risk for cancer progression and selection of additional therapy. METHODS: We performed deep sequencing (CAncer Personalized Profiling by deep Sequencing, [CAPP-Seq]) analyses of plasma cell-free DNA collected from 45 patients before and after chemoradiotherapy for esophageal cancer, as well as DNA from leukocytes and fixed esophageal tumor biopsy samples collected during esophagogastroduodenoscopy. Patients were treated from May 2010 through October 2015; 23 patients subsequently underwent esophagectomy, and 22 did not undergo surgery. We also sequenced DNA from blood samples from 40 healthy control individuals. We analyzed 802 regions of 607 genes for single-nucleotide variants previously associated with esophageal adenocarcinoma or squamous cell carcinoma. Patients underwent imaging analyses 6-8 weeks after chemoradiotherapy and were followed for 5 years. Our primary aim was to determine whether detection of circulating tumor DNA (ctDNA) after chemoradiotherapy is associated with risk of tumor progression (growth of local, regional, or distant tumors, detected by imaging or biopsy). RESULTS: The median proportion of tumor-derived DNA in total cell-free DNA before treatment was 0.07%, indicating that ultrasensitive assays are needed for quantification and analysis of ctDNA from localized esophageal tumors. Detection of ctDNA after chemoradiotherapy was associated with tumor progression (hazard ratio, 18.7; P < .0001), formation of distant metastases (hazard ratio, 32.1; P < .0001), and shorter disease-specific survival times (hazard ratio, 23.1; P < .0001). A higher proportion of patients with tumor progression had new mutations detected in plasma samples collected after chemoradiotherapy than patients without progression (P = .03). Detection of ctDNA after chemoradiotherapy preceded radiographic evidence of tumor progression by an average of 2.8 months. Among patients who received chemoradiotherapy without surgery, combined ctDNA and metabolic imaging analysis predicted progression in 100% of patients with tumor progression, compared with 71% for only ctDNA detection and 57% for only metabolic imaging analysis (P < .001 for comparison of either technique to combined analysis). CONCLUSIONS: In an analysis of cell-free DNA in blood samples from patients who underwent chemoradiotherapy for esophageal cancer, detection of ctDNA was associated with tumor progression, metastasis, and disease-specific survival. Analysis of ctDNA might be used to identify patients at highest risk for tumor progression.

MicroRNA-34a perturbs B lymphocyte development by repressing the forkhead box transcription factor Foxp1.

MicroRNAs (miRNAs) can influence lineage choice or affect critical developmental checkpoints during hematopoiesis. We examined the role of the p53-induced microRNA miR-34a in hematopoiesis by gain-of-function analysis in murine bone marrow. Constitutive expression of miR-34a led to a block in B cell development at the pro-B-cell-to-pre-B-cell transition, leading to a reduction in mature B cells. This block appeared to be mediated primarily by inhibited expression of the transcription factor Foxp1. Foxp1 was a direct target of miR-34a in a 3'-untranslated region (UTR)-dependent fashion. Knockdown of Foxp1 by siRNA recapitulated the B cell developmental phenotype induced by miR-34a, whereas cotransduction of Foxp1 lacking its 3' UTR with miR-34a rescued B cell maturation. Knockdown of miR-34a resulted in increased amounts of Foxp1 and mature B cells. These findings identify a role for miR-34a in connecting the p53 network with suppression of Foxp1, a known B cell oncogene.

MicroRNA-155 promotes autoimmune inflammation by enhancing inflammatory T cell development.

Mammalian noncoding microRNAs (miRNAs) are a class of gene regulators that have been linked to immune system function. Here, we have investigated the role of miR-155 during an autoimmune inflammatory disease. Consistent with a positive role for miR-155 in mediating inflammatory responses, Mir155(-/-) mice were highly resistant to experimental autoimmune encephalomyelitis (EAE). miR-155 functions in the hematopoietic compartment to promote the development of inflammatory T cells including the T helper 17 (Th17) cell and Th1 cell subsets. Furthermore, the major contribution of miR-155 to EAE was CD4(+) T cell intrinsic, whereas miR-155 was also required for optimum dendritic cell production of cytokines that promoted Th17 cell formation. Our study shows that one aspect of miR-155 function is the promotion of T cell-dependent tissue inflammation, suggesting that miR-155 might be a promising therapeutic target for the treatment of autoimmune disorders.

Dual mechanisms by which miR-125b represses IRF4 to induce myeloid and B-cell leukemias.

The oncomir microRNA-125b (miR-125b) is upregulated in a variety of human neoplastic blood disorders and constitutive upregulation of miR-125b in mice can promote myeloid and B-cell leukemia. We found that miR-125b promotes myeloid and B-cell neoplasm by inducing tumorigenesis in hematopoietic progenitor cells. Our study demonstrates that miR-125b induces myeloid leukemia by enhancing myeloid progenitor output from stem cells as well as inducing immortality, self-renewal, and tumorigenesis in myeloid progenitors. Through functional and genetic analyses, we demonstrated that miR-125b induces myeloid and B-cell leukemia by inhibiting interferon regulatory factor 4 (IRF4) but through distinct mechanisms; it induces myeloid leukemia through repressing IRF4 at the messenger RNA (mRNA) level without altering the genomic DNA and induces B-cell leukemia via genetic deletion of the gene encoding IRF4.

MicroRNA-155 confers encephalogenic potential to Th17 cells by promoting effector gene expression.

Th17 cells are central to the pathogenesis of autoimmune disease, and recently specific noncoding microRNAs have been shown to regulate their development. However, it remains unclear whether microRNAs are also involved in modulating Th17 cell effector functions. Consequently, we examined the role of miR-155 in differentiated Th17 cells during their induction of experimental autoimmune encephalomyelitis. Using adoptive transfer experiments, we found that highly purified, myelin oligodendrocyte glycoprotein Ag-specific Th17 cells lacking miR-155 were defective in their capacity to cause experimental autoimmune encephalomyelitis. Gene expression profiling of purified miR-155(-/-)IL-17F(+) Th17 cells identified a subset of effector genes that are dependent on miR-155 for their proper expression through a mechanism involving repression of the transcription factor Ets1. Among the genes reduced in the absence of miR-155 was IL-23R, resulting in miR-155(-/-) Th17 cells being hyporesponsive to IL-23. Taken together, our study demonstrates a critical role for miR-155 in Th17 cells as they unleash autoimmune inflammation and finds that this occurs through a signaling network involving miR-155, Ets1, and the clinically relevant IL-23-IL-23R pathway.

Oncomir miR-125b regulates hematopoiesis by targeting the gene Lin28A.

MicroRNA-125b (miR-125b) is up-regulated in patients with leukemia. Overexpression of miR-125b alone in mice causes a very aggressive, transplantable myeloid leukemia. Before leukemia, these mice do not display elevation of white blood cells in the spleen or bone marrow; rather, the hematopoietic compartment shows lineage-skewing, with myeloid cell numbers dramatically increased and B-cell numbers severely diminished. miR-125b exerts this effect by up-regulating the number of common myeloid progenitors while inhibiting development of pre-B cells. We applied a miR-125b sponge loss of function system in vivo to show that miR-125b physiologically regulates hematopoietic development. Investigating the mechanism by which miR-125b regulates hematopoiesis, we found that, among a panel of candidate targets, the mRNA for Lin28A, an induced pluripotent stem cell gene, was most repressed by miR-125b in mouse hematopoietic stem and progenitor cells. Overexpressing Lin28A in the mouse hematopoietic system mimicked the phenotype observed on inhibiting miR-125b function, leading to a decrease in hematopoietic output. Relevant to the miR-125b overexpression phenotype, we also found that knockdown of Lin28A led to hematopoietic lineage-skewing, with increased myeloid and decreased B-cell numbers. Thus, the miR-125b target Lin28A is an important regulator of hematopoiesis and a primary target of miR-125b in the hematopoietic system.

Racial Disparities in Early-Stage NSCLC Treatment: A Call for Action.

Non-small cell lung cancer (NSCLC) is associated with a 5-year survival rate of only 28%; however, when caught at an early stage, it can be cured with surgery or stereotactic body radiotherapy (SBRT). Unfortunately, racial disparities may result in limited access to care for some patients. Liu and colleagues analyzed 64,999 cases of early-stage NSCLC treated between 2005 and 2017 from the Florida Cancer Registry and showed that Black patients had 36% lower odds of receiving curative-intent surgery compared with their White counterparts. This study highlights significant racial disparities in treatment patterns that must be addressed urgently. See related article by Liu and colleagues, Cancer Epidemiol Biomarkers Prev 2024;33:489-99.

Mapping single-cell developmental potential in health and disease with interpretable deep learning.

Single-cell RNA sequencing (scRNA-seq) has transformed our understanding of cell fate in developmental systems. However, identifying the molecular hallmarks of potency - the capacity of a cell to differentiate into other cell types - has remained challenging. Here, we introduce CytoTRACE 2, an interpretable deep learning framework for characterizing potency and differentiation states on an absolute scale from scRNA-seq data. Across 31 human and mouse scRNA-seq datasets encompassing 28 tissue types, CytoTRACE 2 outperformed existing methods for recovering experimentally determined potency levels and differentiation states covering the entire range of cellular ontogeny. Moreover, it reconstructed the temporal hierarchy of mouse embryogenesis across 62 timepoints; identified pan-tissue expression programs that discriminate major potency levels; and facilitated discovery of cellular phenotypes in cancer linked to survival and immunotherapy resistance. Our results illuminate a fundamental feature of cell biology and provide a broadly applicable platform for delineating single-cell differentiation landscapes in health and disease.