Stanford Advisors


All Publications


  • Discovery of sparse, reliable omic biomarkers with Stabl. Nature biotechnology Hédou, J., Marić, I., Bellan, G., Einhaus, J., Gaudillière, D. K., Ladant, F. X., Verdonk, F., Stelzer, I. A., Feyaerts, D., Tsai, A. S., Ganio, E. A., Sabayev, M., Gillard, J., Amar, J., Cambriel, A., Oskotsky, T. T., Roldan, A., Golob, J. L., Sirota, M., Bonham, T. A., Sato, M., Diop, M., Durand, X., Angst, M. S., Stevenson, D. K., Aghaeepour, N., Montanari, A., Gaudillière, B. 2024

    Abstract

    Adoption of high-content omic technologies in clinical studies, coupled with computational methods, has yielded an abundance of candidate biomarkers. However, translating such findings into bona fide clinical biomarkers remains challenging. To facilitate this process, we introduce Stabl, a general machine learning method that identifies a sparse, reliable set of biomarkers by integrating noise injection and a data-driven signal-to-noise threshold into multivariable predictive modeling. Evaluation of Stabl on synthetic datasets and five independent clinical studies demonstrates improved biomarker sparsity and reliability compared to commonly used sparsity-promoting regularization methods while maintaining predictive performance; it distills datasets containing 1,400-35,000 features down to 4-34 candidate biomarkers. Stabl extends to multi-omic integration tasks, enabling biological interpretation of complex predictive models, as it hones in on a shortlist of proteomic, metabolomic and cytometric events predicting labor onset, microbial biomarkers of pre-term birth and a pre-operative immune signature of post-surgical infections. Stabl is available at https://github.com/gregbellan/Stabl .

    View details for DOI 10.1038/s41587-023-02033-x

    View details for PubMedID 38168992

    View details for PubMedCentralID 7003173

  • Spatial subsetting enables integrative modeling of oral squamous cell carcinoma multiplex imaging data. iScience Einhaus, J., Gaudilliere, D. K., Hedou, J., Feyaerts, D., Ozawa, M. G., Sato, M., Ganio, E. A., Tsai, A. S., Stelzer, I. A., Bruckman, K. C., Amar, J. N., Sabayev, M., Bonham, T. A., Gillard, J., Diop, M., Cambriel, A., Mihalic, Z. N., Valdez, T., Liu, S. Y., Feirrera, L., Lam, D. K., Sunwoo, J. B., Schürch, C. M., Gaudilliere, B., Han, X. 2023; 26 (12): 108486

    Abstract

    Oral squamous cell carcinoma (OSCC), a prevalent and aggressive neoplasm, poses a significant challenge due to poor prognosis and limited prognostic biomarkers. Leveraging highly multiplexed imaging mass cytometry, we investigated the tumor immune microenvironment (TIME) in OSCC biopsies, characterizing immune cell distribution and signaling activity at the tumor-invasive front. Our spatial subsetting approach standardized cellular populations by tissue zone, improving feature reproducibility and revealing TIME patterns accompanying loss-of-differentiation. Employing a machine-learning pipeline combining reliable feature selection with multivariable modeling, we achieved accurate histological grade classification (AUC = 0.88). Three model features correlated with clinical outcomes in an independent cohort: granulocyte MAPKAPK2 signaling at the tumor front, stromal CD4+ memory T cell size, and the distance of fibroblasts from the tumor border. This study establishes a robust modeling framework for distilling complex imaging data, uncovering sentinel characteristics of the OSCC TIME to facilitate prognostic biomarkers discovery for recurrence risk stratification and immunomodulatory therapy development.

    View details for DOI 10.1016/j.isci.2023.108486

    View details for PubMedID 38125025

    View details for PubMedCentralID PMC10730356

  • Impact of preoperative uni- or multimodal prehabilitation on postoperative morbidity: meta-analysis. BJS open Cambriel, A., Choisy, B., Hedou, J., Bonnet, M., Fellous, S., Lefevre, J. H., Voron, T., Gaudilliere, D., Kin, C., Gaudilliere, B., Verdonk, F. 2023; 7 (6)

    Abstract

    BACKGROUND: Postoperative complications occur in up to 43% of patients after surgery, resulting in increased morbidity and economic burden. Prehabilitation has the potential to increase patients' preoperative health status and thereby improve postoperative outcomes. However, reported results of prehabilitation are contradictory. The objective of this systematic review is to evaluate the effects of prehabilitation on postoperative outcomes (postoperative complications, hospital length of stay, pain at postoperative day 1) in patients undergoing elective surgery.METHODS: The authors performed a systematic review and meta-analysis of RCTs published between January 2006 and June 2023 comparing prehabilitation programmes lasting ≥14 days to 'standard of care' (SOC) and reporting postoperative complications according to the Clavien-Dindo classification. Database searches were conducted in PubMed, CINAHL, EMBASE, PsycINFO. The primary outcome examined was the effect of uni- or multimodal prehabilitation on 30-day complications. Secondary outcomes were length of ICU and hospital stay (LOS) and reported pain scores.RESULTS: Twenty-five studies (including 2090 patients randomized in a 1:1 ratio) met the inclusion criteria. Average methodological study quality was moderate. There was no difference between prehabilitation and SOC groups in regard to occurrence of postoperative complications (OR = 1.02, 95% c.i. 0.93 to 1.13; P = 0.10; I2 = 34%), total hospital LOS (-0.13 days; 95% c.i. -0.56 to 0.28; P = 0.53; I2 = 21%) or reported postoperative pain. The ICU LOS was significantly shorter in the prehabilitation group (-0.57 days; 95% c.i. -1.10 to -0.04; P = 0.03; I2 = 46%). Separate comparison of uni- and multimodal prehabilitation showed no difference for either intervention.CONCLUSION: Prehabilitation reduces ICU LOS compared with SOC in elective surgery patients but has no effect on overall complication rates or total LOS, regardless of modality. Prehabilitation programs need standardization and specific targeting of those patients most likely to benefit.

    View details for DOI 10.1093/bjsopen/zrad129

    View details for PubMedID 38108466

  • Impact of Preoperative Uni- or Multimodal Prehabilitation on Postoperative Morbidity: A Systematic Review and Meta-Analysis Verdonk, F., Choisy, B., Cambriel, A., Hedou, J., Bonnet, M., Lefevre, J., Voron, T., Gaudilliere, D., Kin, C., Gaudilliere, B. LIPPINCOTT WILLIAMS & WILKINS. 2023: 806-807
  • An Immune Signature of Surgical Site Infections (SSI), a Retrospective Study with a Novel Machine Learning Pipeline for Biomarker Identification Verdonk, F., Hedou, J., Maric, I., Bellan, G., Einhaus, J., Gaudilliere, D., Bonham, A., Angst, M., Gaudilliere, B., Cambriel, A. LIPPINCOTT WILLIAMS & WILKINS. 2023: 773-774
  • Risk factors for prolonged time to hospital discharge after ambulatory cholecystectomy under general anaesthesia. A retrospective cohort study. International journal of surgery (London, England) Picard, L., Duceau, B., Cambriel, A., Voron, T., Makoudi, S., Tsai, A., Yazid, L., Soulier, A., Paugam, C., Lescot, T., Bonnet, F., Verdonk, F. 2022: 106706

    Abstract

    BACKGROUND: Although predictive models have already integrated demographic factors and comorbidities as risk factors for a prolonged hospital stay, factors related to anaesthesia management in ambulatory surgery have not been yet characterized. This study aims to identify anaesthetic factors associated with a prolonged discharge time in ambulatory surgery.METHODS: All clinical records of patients who underwent ambulatory cholecystectomy in a French University Hospital (Hopital Saint Antoine, Paris) between January 1st, 2012 and December 31st, 2018 were retrospectively reviewed. The primary endpoint was the discharge time, defined as the time between the end of surgery and discharge. A multivariable Cox proportional-hazards model was fitted to investigate the factors associated with a prolonged discharge time.RESULTS: Five hundred and thirty-five (535) patients were included. The median time for discharge was 150 min (interquartile range - IQR [129-192]). A bivariable analysis highlighted a positive correlation between discharge timeline and the doses-weight of ketamine and sufentanil. In the multivariable Cox proportional hazards model analysis, the anaesthesia-related factors independently associated with prolonged discharge time were the dose-weight of ketamine in interaction with the dose weight of sufentanil (HR 0.10 per increment of 0.1 mg/kg of ketamine or 0.2 mug/kg of sufentanil, CI 95% [0.01-0.61], p = 0.013) and the non-use of a non-steroidal anti-inflammatory drug (NSAID) (HR 0.81 [0.67-0.98], p = 0.034). Twenty patients (4%) had unscheduled hospitalization following surgery.CONCLUSION: Anaesthesia management, namely the use of ketamine and the non-use of NSAID, affects time to hospital discharge.

    View details for DOI 10.1016/j.ijsu.2022.106706

    View details for PubMedID 35697325