Virtual Library

Start Your Search

D.S. Tan



Author of

  • +

    MO10 - Molecular Pathology II (ID 127)

    • Event: WCLC 2013
    • Type: Mini Oral Abstract Session
    • Track: Pathology
    • Presentations: 1
    • +

      MO10.06 - Intertumour heterogeneity revealed by integrative analysis of targeted somatic mutation profiling and whole genome copy-number alterations in non-squamous non-small cell lung carcinomas (ID 3386)

      16:45 - 16:50  |  Author(s): D.S. Tan

      • Abstract
      • Presentation
      • Slides

      Background
      Molecular subtypes of NSCLC are delineated through single driver alterations, although this likely underestimates the extent of inter-tumor heterogeneity. The goal of this study was to evaluate the potential impact of co-occuring mutations and copy number alterations in a series of tumours from both East-Asian (EA) and Western-Europe (WE) origins.

      Methods
      230 non-squamous non-small cell lung carcinomas (NSCLC) were analysed using the MassARRAY LungCarta™ panel (Sequenom) which examines 214 mutations in 26 oncogenes and tumour suppressor genes. Results were integrated with copy number alterations evaluated using Affymetrix Genome-Wide Human SNP 6.0 arrays and clinical variables.

      Results
      Out of the 230 tumours tested, 185 mutations were observed, with 138 tumours (60%) with at least one mutation – 97 tumours (42.2%) with a single mutation in a single gene, 36 tumours (15.7%) with two mutations (either in the same gene or two different genes), four tumours (2.2%) with three mutations and one with four mutations. The most frequent mutations were EGFR (38.5% EA;10.1% WE), KRAS (10.9% EA;31.6% WE), P53 (8.8% EA;15.1% WE), MET (14.3% EA;2.2% WE), STK11 (9.9% EA;2.8% WE) and PIK3CA (2.2% EA;4.3% WE). Co-occurring mutations were found in up to 30% of tumors, although the likelihood differed for each gene: KRAS (16/54, 31%), EGFR (18/49, 37%), MET (6/16, 38%), P53 (13/29, 45%), STK11 (8/13, 62%) and PIK3CA (5/8; 62%). Among the eight tumors harbouring PIK3CA mutations, five cases had a co-mutation (four cases with EGFR, one case with KRAS). In EGFR mutant cancers, co-occuring mutations include p53 (10%), PIK3CA (8%), STK11 (6%) and MET (4%). Significant relationships were also detected between EGFR mutation and CNAs on chromosomes 1p, 7p, and 13q. There were also significant relationships between KRAS mutation and CNAs on chromosomes 1q and 3q. For stage I, we found a worse prognosis for patients with at least one mutation. PIK3CA was significantly correlated with poor prognosis. Consistent with recent pooled analysis, KRAS alone is not prognostic but when CNAs and mutation status are combined, patients having both KRAS mutation and the highest related CNA (3q22.3 copy loss) showed a significant poorer prognosis.

      Conclusion
      This study highlights the wide diversity of mutation profiles within molecularly-defined NSCLCs, revealing co-mutations and associations with numerical chromosomal abnormalities that are clinically relevant. This diversity should be taken into account when designing stratified treatment approaches and underscores the need for customized assays that broadly screen for “actionable” mutations and copy-number alterations.

      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.

  • +

    O03 - NSCLC - Targeted Therapies I (ID 113)

    • Event: WCLC 2013
    • Type: Oral Abstract Session
    • Track: Medical Oncology
    • Presentations: 1
    • +

      O03.03 - Outcomes of NSCLC patients on Phase I Trials: The Importance of Molecular and Patient Selection (ID 3349)

      10:50 - 11:00  |  Author(s): D.S. Tan

      • Abstract
      • Presentation
      • Slides

      Background
      With recent successes of targeted therapeutics, there has been increased enthusiasm to enroll patients with NSCLC into phase I trials. Most phase I prognostic indices have been derived from patient populations where NSCLC is underrepresented.

      Methods
      Retrospective review of NSCLC patients enrolled between 2005-2012 at National Cancer Centre Singapore was performed, collating data on demographics, molecular profiles, trial characteristics and clinical outcomes (using RECIST criteria) to identify prognostic factors to improve patient selection.

      Results
      167 patients were treated on 20 phase I trials. Median (range) age 60.7 years (22.7–84.7), 58% male, adenocarcinoma/squamous/NOS/others (75.4%/9.6%/12.6%/2.4%). 13% were at high risk of nutritional deficiency (BMI<18.5), and ECOG 0(21%)/1(77%)/2(2.4%). Median (range) prior chemotherapy was 2 (0-6) and 99% received at least one treatment line. Class of agents include anti-angiogenics (50.3%), signal transduction pathway inhibitors (STPi) 46%. Only 4.2% received cytotoxics alone. 86.2% were on combination regimens, of which two-thirds on combinations of chemotherapy and small molecule. Between 2008-2012, proportion of patient tumors with molecular alterations identified increased from 5.5% to 62.5%, facilitating enrolment into trials designed for a specific genotype (or “matched” trials). 35% and 7.8% of patients participated on trials targeting EGFR and ALK alterations respectively. With a median follow up of 8.7 months, progression free survival was 3.9m (95%CI:3.4–5.8), overall survival (OS) 10.4m (95%CI:8.0–11.7). Among evaluable patients, 24.8% (95%CI:18.2–32.5) had complete/partial response; clinical benefit rate (CBR) 75.8% (95%CI:68.2–82.4). 90-day mortality (90DM) was 15%. Patients participating in “matched” trials had a lower risk of death (HR 0.55 [95%CI:0.38-0.78] p<0.001) compared to those on unselected or “unmatched” trials (median OS 11.9 v 7.6). On univariate analysis, low BMI<18.5, ECOG>0, ≥3 metastatic sites, presence of bone metastasis, low albumin (<35), low Hb (<12), ≥ 3 prior lines of chemotherapy and non-participation in a "matched" trial were significant negative prognostic factors for OS. On multivariate analysis, ECOG, number of metastatic sites, albumin and trial type (targeting EGFR/ALK/others), emerged as independent variables. These factors were used to construct a prognostic nomogram to predict OS at 3, 6 and 12m. Of 12 published prognostic indices, 8 models were validated in our patient cohort, with the highest c-index being 0.66. Pairwise comparison against these 8 (7 tumor agnostic; 1 NSCLC) prognostic indices, found the nomogram to be superior in predicting OS, with a c-index of 0.74.

      Conclusion
      This is the largest analysis of phase I NSCLC patients using individual patient data. Outcome of NSCLC patients in phase I trials is promising with CBR 76% and median OS 10.8m. Our nomogram – comprising of ECOG, albumin, number of metastatic sites, participation in a “matched” trial – is uniquely derived from NSCLC patients and was a better predictor of OS compared to 8 published phase I prognostic scores.

      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.