Virtual Library

Start Your Search

T. Berghmans



Author of

  • +

    O27 - Clinical Trials and Practice (ID 142)

    • Event: WCLC 2013
    • Type: Oral Abstract Session
    • Track: Other Topics
    • Presentations: 1
    • +

      O27.05 - Is primary tumor standardized uptake value (SUV) an independent prognostic factor for non-small cell lung cancer (NSCLC)? A meta-analysis based on individual patients data. (ID 3888)

      16:15 - 17:45  |  Author(s): T. Berghmans

      • Abstract
      • Presentation
      • Slides

      Background
      [18]F-fluoro-2-deoxy-D-glucose positron emission tomography complements conventional imaging for staging lung cancer although its ability to predict outcome is less well established. Two literature-based meta-analyses suggest a prognostic value in univariate analysis. To assess FDG-PET value in predicting survival adjusted for some known prognostic factors, we carried out a meta-analysis based on individual patients data from multiple independent studies.

      Methods
      Following literature search, and after writing of a protocol for the meta-analysis, we contacted the authors of identified studies and requested individual patients data; we also tried to collect some unpublished data. Data analysis used Cox regression models stratified for the study with overall survival as primary outcome. SUV max was used as a binary covariate (median value for each study).

      Results
      Data were collected for 1526 patients (57% of the identified patients) from 11 publications and 1 unpublished series (median age : 64 years, 60% male patients, squamous cell in 34%, adenocarcinoma in 47%, stages I-II in 58%). Combined univariate hazard ratio (HR) was 1.43 (95% CI : 1.22-1.66); no statistically significant interaction between SUV and one of six additional freatures (age, gender, histology,stage, tumors size –in stages I-III patients- and surgical treatment), was found except for stage (p=0.05) with a decreased prognostic value of SUV for stage IV patients. Without considering SUV, multivariate analysis identified, in stage I-III patients, age, stage, tumor size and surgical treatment as independent prognostic factors. The addition of SUV improved that model : HR estimate for SUV effect was 1.58, statistically significant (95% CI : 1.27-1.96), p<0.0001. No interaction was found with SUV. When tumor size was not included in the tested covariates, we found SUV of additional value (adjustment for age, stage, surgical treatment) with a HR of 1.35 (95% CI : 1.15-158). Interaction between SUV and stage was detected, restricting the significant impact of SUV on survival to stage I-III patients.

      Conclusion
      Conclusions : Although suffering from selection bias and lack of homogeneous SUV assessment, these data suggest that SUV at the time of diagnosis is an independent prognostic marker for patients with stage I-III NSCLC. The utility of SUV in predicting survival in stage IV patients requires further studies.

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

  • +

    O28 - Endoscopy (ID 124)

    • Event: WCLC 2013
    • Type: Oral Abstract Session
    • Track: Pulmonology + Endoscopy/Pulmonary
    • Presentations: 1
    • +

      O28.05 - An analysis of a prospective study from the European Lung Cancer Working Party (ELCWP) looking at predictive factors for response to chemotherapy (CT): limitations in translational cooperative research. (ID 1941)

      10:30 - 12:00  |  Author(s): T. Berghmans

      • Abstract
      • Presentation
      • Slides

      Background
      Adequate tumour samplings for biological analyses are currently of major importance in treating oncological patients. Obtaining histological samplings from the primary lung cancer can be a challenge due to tumour accessibility, small biopsies or tolerance to bronchoscopy such as bleeding or dyspnoea in case of limited airflow capacity. The ELCWP developed a multicentre prospective study searching for predictive factors for response to chemotherapy based on genomic analyses. We aim to analyse the capability in obtaining adequate tumour samplings from the primary non small cell lung cancer (NSCLC) for studying the transcriptome (miRNA and mRNAs) with high throughput techniques.

      Methods
      All patients presenting with a suspected lung cancer were proposed participating to the study. To be evaluable for the primary endpoint of the study, patients needed to have a confirmed diagnosis of NSCLC treated with chemotherapy and assessable for response. During the diagnostic bronchoscopy, 3 biopsies were collected from the primary tumour, with a control sample from normally appearing bronchial mucosa. One was formalin fixed and paraffin embedded for pathological diagnosis. A second was used for transcriptome analysis and the third one was frozen and stored in a tissue bank. We are presenting the flow chart of the patients screened for entry in the ELCWP study and the limitations for obtaining tumour samplings in assessable patients.

      Results
      From 1/04/2009 to 12/06/2013, 307 patients suspected to have NSCLC were prospectively registered. Eleven are under evaluation for pending histological confirmation leaving 296 patients evaluable for the present analysis. In 25 cases, no lung cancer confirmation was obtained (other tumour n = 12, no pathological confirmation at all n = 6, benign lesion n = 6, other reason n = 1) and 6 further patients withdrew their initial consent. Among 265 pathologically confirmed lung cancer (samples obtained during bronchoscopy or by another technique), 38 small cell lung cancers (SCLC) and 227 NSCLC were diagnosed. In addition to the diagnostic biopsy, further samplings for genomic analyses could be obtained during the same bronchoscopy in 30/38 SCLC (79%) and 116/227 NSCLC (51%). Among 227 NSCLC, 107 were presenting with an advanced disease treated with a cisplatin-based chemotherapy and were assessable for response to chemotherapy (primary study endpoint). Among these 107 patients, 59 adequate tumour samplings could be obtained for transcriptome analysis (20% from the initial cohort and 55% among assessable patients).

      Conclusion
      This analysis of a prospective multicentre study is showing the difficulties and limitations in obtaining adequate tumour samplings for biological analyses when conducting translational cooperative research in non-small cell lung cancer.

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

  • +

    P2.24 - Poster Session 2 - Supportive Care (ID 157)

    • Event: WCLC 2013
    • Type: Poster Session
    • Track: Supportive Care
    • Presentations: 1
    • +

      P2.24-021 - Adjuvant or Induction chemotherapy for Non Small Cell Lung Cancer treated with chemoradiotherapy: An invidual data metaanalysis of phase II trials (ID 1421)

      09:30 - 16:30  |  Author(s): T. Berghmans

      • Abstract

      Background
      it is well known that combining chemotherapy and radiation therapy is beneficial to patients with locally advanced non small cell lung cancer compared to radiation alone or compared to a sequential approach using chemotherapy and radiation therapy. However, it is not obvious what is the best schedule. A few randomized trials assessed chemotherapy as induction before chemoradiotherapy (CT -> CTRT) versus chemotherapy as consolidation, after chemoradiotherapy (CTRT -> CT). Most of those trials are phase II trials with moderate sample sizes and were not designed to demonstrate treatment effect in terms of overall survival.

      Methods
      the study coordinators of those trials (T. Berghmans, H. Choy, P. Fournel, P. Garrido, J. Van Meerbeeck) agreed on a protocol for carrying out a meta-analysis of individual patients data and for sharing the individual patients data that were sent to the coordinating institution. Overall survival was the primary outcome, progression-free survival and toxic death occurrence were among the secondary outcomes. The treatment effect was assessed through the estimation of the hazard ratio of the survival distributions using CTRT -> CT as reference. Combined hazard ratio was obtained through Cox regression models (fixed effects) with a stratification by trial. Preplanned interactions between baseline covariates (age, sex, performance status, stage, histology) and treatment effect were assessed. Toxic death rates were analyzed per trial and odds ratios have been estimated to assess the treatment effect. Combined odds ratio was obtained by the Peto method.

      Results
      the data bases of the 5 eligible identified trials (3 with cisplatin based chemotherapy regimens, 2 with carboplatin based regimens) were shared for a total of 534 patients (CT -> CTRT 271, CTRT -> CT 263). Median ages were 60 and 61 years, stage IIIB represented 69%/70% of the patients and EOCG PS > 1 was rare (3%/2%). Median follow-up ranged from 12 months up to 66 months and rates of events from 44% to 88%. No significant difference was detected either for overall survival with an estimated HR of 0.96 (95% CI : 0.79-1.17) without heterogeneity between the 5 trials (I[2]=0) or for progression-free survival (analysis restricted to 4 out of the 5 trials), HR=0.91 (95% CI : 0.75-1.11) and absence of heterogeneity (I[2]=2%). For both outcomes, no interaction between the above specified covariates and treatment effect was found. Toxic deaths occurred overall in 3% of the patients, no detectable impact of treatment arm was found with a combined odds ratio of 0.40 and a 95 % CI overlapping 1 (0.15-1.06).

      Conclusion
      our results suggest that there is no argument in favour of one of the two therapeutic schedules when looking at overall survival or at progression free survival; however, in the absence of benefit in terms of prognosis, a more detailed evaluation of toxicity is warranted and is ongoing.