MétaCan
Menu
Back to cohort
Record W6997061831

A transcriptomic analysis of intratumor and stromal heterogeneity in breast cancer

2017· dissertation· en· W6997061831 on OpenAlexfundno aff

Bibliographic record

VenueeScholarship@McGill (McGill) · 2017
Typedissertation
Languageen
FieldBiochemistry, Genetics and Molecular Biology
TopicSingle-cell and spatial transcriptomics
Canadian institutionsnot available
FundersCanadian Institutes of Health ResearchFaculty of Medicine, McGill UniversityMcGill University
KeywordsSubtypingStromal cellBreast cancerTranscriptomeStromaTriple-negative breast cancerGene expression profilingTumor microenvironment
DOInot available

Abstract

fetched live from OpenAlex

The management of breast cancer is complicated by inter-and intra-tumour heterogeneity.In particular, triple negative breast cancer (TNBC) is a difficult to treat, molecularly heterogeneous cancer subtype that lacks actionable targets.The heterogeneity of the TNBC microenvironment (stroma) has not been well characterized despite the key role that it may play in tumor progression.Similarly, the impact of intra-tumoral heterogeneity on therapeutic response and patient outcome remains largely unknown.To address these challenges I investigated the transcriptome of tumor-associated stroma isolated from TNBCs (n=57), as well as comprehensive single-cell gene expression profiling from a treatment-resistant breast patient-derived xenograft (PDX) displaying heterogeneity for the therapeutically targetable HER2 receptor (n=33 cells).Analysis of the TNBC stroma identified four stromal properties associated with T cells (T), B cells (B), invasive epithelial cells (E), or a desmoplastic reaction (D) respectively.My method, entitled STROMA4, assigns each sample as either low, intermediate, or high for each property independently in stromal or bulk expression profiles.I provide evidence that TNBCType, a previously reported subtyping scheme for TNBC, underestimates the complexity of some tumors, and show how stratification by the STROMA4 method can predict patient benefit from therapy with increased sensitivity.Combining the STROMA4 property assignments generates a novel TNBC subtyping scheme, and analysis of this subtyping scheme revealed that only 15 of 81 possible subtypes had larger than expected populations.This combinatorial IV approach revealed that the B, T and E properties are prognostic only when the D property is not high, providing a potential explanation for misprediction by existing classifiers.Analysis of single-cell RNA-seq (scRNA-seq) data from a PDX heterogeneous for HER2 expression identified distinct cellular subpopulations and revealed a predominantly basal breast cancer subtype.Unsupervised hierarchical clustering distinguished two major cellular subpopulations with differential expression of EGFR, which was validated immunohistochemically in the resected tumour.Further investigation into differences between the EGFR-high and -low cells in the scRNA-seq data indicated that EGFR-high cells were more "stem-like", which was then validated experimentally.The presence of EGFR-high stem cells in this PDX model, as well as in other PDX models, is associated with sensitivity to EGFR inhibition.Analysis of the TNBC stroma, using a multi-parameter classification model, produces a simple ontology that captures TNBC heterogeneity, and informs how tumor-associated properties and biologies interact to affect prognosis; while analysis of the scRNA-seq data identified two groups of cells with differential expression of EGFR and stem-like characteristics, which is associated with response to EGFR inhibition.Thus, this work adds to our understanding of the contribution of inter-and intra-tumoral heterogeneity to the complexity of the cancer ecosystem, and the effect it has on response to therapy.V RSUM L'htrognit inter et intra-tumorale participe la complexit de la biologie du cancer du sein.Plus particulirement, le cancer du sein triple ngatif (CSTN) est difficile traiter en raison de son htrognit au niveau molculaire et manque de cibles thrapeutiques actionnables.L'htrognit du microenvironnement tumoral des CSTN reste peu caractrise malgr le rle cl que ce dernier peut jouer dans la progression tumorale.De faon similaire, l'impact de l'htrognit intra-tumorale sur la rponse thrapeutique et la survie des patients demeure inconnue.Afin d'lucider ces mcanismes, j'ai analys le transcriptome du stroma tumoral isol partir de CSTN (n=57) ainsi que le profil d'expression cellulaire (cellules individuelles ; n=33) d'une xnogreffe drive de tumeur de patient (XDP) rsistante la thrapie et arborant une htrognit d'expression du rcepteur HER2.Ce rcepteur peut tre cibl de faon thrapeutique en clinique.L'analyse du stroma des CSTN a permis d'identifier quatre proprits stromales associes aux cellules T (T), B (B), aux cellules pithliales invasives pithliales (E), ou une raction desmoplasique (D).La mthode que j'ai dveloppe, intitule STROMA4 , assigne un score faible, intermdiaire ou lev pour chaque proprit partir des profils d'expression gnique du stroma (stroma tumoral) ou de la tumeur globale (tumeur en entier).J'ai pu montrer que le TNBCType , une mthode de sous-typage des CSTN ayant pralablement t publie, VI sous-estime la complexit de certaines tumeurs.De plus, la stratification des patientes selon la mthode STROMA4 permet de prdire la rponse la thrapie avec une meilleure sensibilit que la mthode TNBCType .La combinaison des diffrentes proprits identifies par la mthode STROMA4 gnre une nouvelle stratification des CSTN.L'analyse de cette nouvelle classification a permis de montrer que seul 15 des 81 sous-types possibles (d'aprs les diffrentes combinaisons de scores des proprits) sont reprsents par une population plus grande qu'attendue.Cette approche combinatoire rvle que les proprits B, T et E sont pronostiques seulement quand la proprit D est de faible score.Ceci pourrait expliquer, en partie du moins, la mauvaise prdiction des classificateurs existants.L'analyse des donnes provenant du squenage ARN de cellules isoles (scRNA-seq) d'une XPD htrogne pour l'expression de HER2 identifie des sous-populations cellulaires distinctes et rvle, de faon dominante, un sous-type basal de cancer de sein.La classification hirarchique ( hierarchical clustering en anglais) non supervise distingue deux souspopulations cellulaires majeures ayant des taux d'expression du rcepteur EGFR diffrentes au niveau de l'expression gnique.Cette diffrence est valide au niveau protique par immunohistochimie sur un chantillon de tumeur humaine rsque au moment de la chirurgie de la patiente.L'tude des diffrences entre les cellules forte et faible expression de EGFR par scRNA-seq indique que les cellules ayant une expression leve de EGFR arborent des proprits de cellules souches .Ce rsultat a t valide de faon exprimentale.La prsence de cellules ayant un fort taux d'expression de EGFR dans ce modle ainsi que dans d'autres modles XPD, est associe une sensibilit vis vis de l'inhibition de EGFR.VII L'analyse du stroma des CSTN, utilisant un modle de classification multi-paramtrique, gnre une classification simple qui rcapitule l'htrognit des CSTN.Cette classification permet de comprendre comment les diffrentes proprits biologiques associes la tumeur interagissent et affectent le pronostic.D'autre part, l'analyse des donnes du scRNA-seq identifie deux groupes de cellules avec des diffrences de (i) niveaux d'expression de EGFR, (ii) proprits de cellules souches .Ces cellules sont galement associes avec une rponse l'inhibition de EGFR.Ainsi, ce travail permet une meilleure comprhension de la contribution de l'htrognit aux niveaux inter-et intra-tumoral la complexit de l'cosystme du cancer et son effet sur la rponse la thrapie.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow)
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Bench or experimental · Consensus signal: Bench or experimental
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.351
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.013
GPT teacher head0.256
Teacher spread0.243 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

Study designBench or experimental
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2017
Admission routes1
Has abstractyes

Explore more

Same venueeScholarship@McGill (McGill)Same topicSingle-cell and spatial transcriptomicsFrench-language works237,207