Genome-Wide Expression Profiling Predicts Treatment Outcome in Classical Hodgkin Lymphoma
Notice bibliographique
Résumé
Abstract INTRODUCTION: Despite advances in Hodgkin lymphoma (HL) treatment about 20% of patients still die due to progressive disease. Current prognostic models predict treatment outcome with imperfect accuracy and clinically relevant biomarkers are yet to be established that improve upon the existing International Prognostic Scoring system. PATIENTS AND METHODS: We analyzed 113 fresh frozen lymph node specimens from classical HL patients by gene expression profiling (Affymetrix UA 133 2.0 Plus) focusing on correlations with treatment outcome. The cohort comprised 100 diagnostic pretreatment and 13 relapse biopsies. Treatment was considered a failure if the lymphoma progressed during therapy or relapsed at any time. Treatment success was defined as absence of progression or relapse. Gene expression findings were validated in paraffin embedded material using immunohistochemistry (IHC) for CD20 and CD56 in the original 100 cases, in an independent validation cohort of 166 cases and a set of 172 cases using flow cytometric assessment of CD20 and CD56. For our predictive model, we trained a classifier by Simultaneous Multinominal Logistic Regression (SMLR) and assessed relative feature importance using Random Forests. RESULTS: We found underexpression of genes representing a global B cell signature in 18 pretreatment biopsies of patients whose first line systemic chemotherapy failed (p=0.033). These results were independently validated using immunohistochemistry for CD20 in 166 patients. The presence of background B cells in the direct vicinity of Hodgkin Reed Sternberg (HRS) cells favorably affected progression-free survival (p=0.042) in a univariate analysis; however, in multivariate analysis only clinical stage and hemoglobin were independent prognostic factors. Furthermore, down-regulation of genes involved in T cell receptor signaling was associated with failure of first line systemic chemotherapy. In contrast to immunohistochemistry validation, flow cytometry was not sensitive to differences in T (CD3) or B cell (CD20) numbers (p=0.22 and p=0.35, respectively). Under-expression of genes of an NK cell signature in 6 relapse biopsies correlated with failure to secondary therapy with autologous stem cell transplantation (p=0.048). IHC accordingly showed complete lack of NK cells in these biopsies. To study the predictive power of gene expression in comparison to clinical risk factors, we identified 103 gene expression probe sets and Ann Arbor stage as the most important features. Importantly, we found 8 probe sets that were more influential than Ann Arbor stage. In comparison to either gene expression or clinical variables alone, combining of the two data sources achieved best performance values for predicting outcome of first line treatment (Receiver Operating Characteristics [ROC]: AUC(combined) = 0.76; AUC(gene expression) = 0.71; AUC(clinical) = 0.69). CONCLUSION: Clinical outcome correlates with the presence of B cells in pretreatment and NK cells in relapse biopsies verifying the importance of the microenvironment for outcome prediction in HL. We were able to validate these findings by IHC and showed that GEP adds to the predictive value of clinical prognostic scoring. Our data suggest that integration of further molecular data, especially those derived from HRS cell enriched specimens, will be of value in helping to inform novel predictive models in Hodgkin lymphoma.
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction machine sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Le volet Gemma est une étiquette directe du modèle pour chaque travail de la base, lue sur la notice réduite au titre. Le volet Codex est un classifieur appris des 10 348 étiquettes directes de Codex et calibré sur les taux pondérés de l'échantillon; les champs sans appui suffisant ne portent aucun appel Codex. Le mode candidate est l'union des deux volets; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont pas des étiquettes humaines.
Scores du classifieur distillé par catégorie (deux têtes)
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,000 | 0,001 |
| Méta-épidémiologie (sens strict) | 0,000 | 0,000 |
| Méta-épidémiologie (sens large) | 0,000 | 0,000 |
| Bibliométrie | 0,000 | 0,000 |
| Études des sciences et des technologies | 0,000 | 0,000 |
| Communication savante | 0,000 | 0,000 |
| Science ouverte | 0,000 | 0,000 |
| Intégrité de la recherche | 0,000 | 0,000 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,001 | 0,000 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule source (Gemma direct ou Codex distillé), pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».