Virtual Library

Start Your Search

M. Finkelman



Author of

  • +

    P1.05 - Poster Session/ Prevention and Tobacco Control (ID 215)

    • Event: WCLC 2015
    • Type: Poster
    • Track: Prevention and Tobacco Control
    • Presentations: 1
    • +

      P1.05-008 - Big Tobacco and the Creation of an Epidemic of Smoking-Related Adenocarcinoma of the Lung: SEER-Based Analysis, 1973-2011 (ID 2479)

      09:30 - 17:00  |  Author(s): M. Finkelman

      • Abstract
      • Slides

      Background:
      When epidemiologic research first demonstrated an association between cigarette smoking and lung cancer in the early 1950s, adenocarcinoma comprised about 5% of lung cancers and appeared to be unrelated to smoking. In the 1960s and 1970s, adenocarcinoma increased sharply, and became strongly related to cigarette smoking. At the 2007 IASLC-sponsored 12th World Conference in Lung Cancer in Seoul, Korea, our group reported that by 2003, adenocarcinoma of the lung had risen to comprise 47% of all lung cancers in the US. The objective of this presentation is to update and expand upon our previous analysis.

      Methods:
      We analyzed time trends in lung cancer histology with changes in cigarette design and Tobacco Industry actions over six decades. We utilized Surveillance-Epidemiology and End Results (SEER) data on 419,941 lung cancers diagnosed between 1973 and 2011 to analyze time trends of age-standardized incidence rates of five histologic subtypes: adenocarcinoma, squamous cell, small cell, large cell, and adenosquamous carcinoma.

      Results:
      Over time, the percentage of lung cancers that were adenocarcinomas increased from 29% (in 1973-1974) to 55% (in 2010-2011). During this 38-year period, the percentage of lung cancers that were squamous cell carcinomas decreased from 41% to 26%. Among all patients, adenocarcinoma incidence surpassed squamous carcinoma by 1985-1989 to become the most common histologic subtype. Adenocarcinoma surpassed squamous cell in 1990-1994 in men, while it was already most common in women by 1973-1974. Adenocarcinoma rose 77% in men from 1973-1974 to 1990-1994, while it rose 197% in women between 1973-1974 and 2005-2006. Among whites, adenocarcinoma surpassed squamous carcinoma by 1985-1989, while this occurred among blacks by 1990-1994. It was already most common among other race individuals in 1973-1974. Adenocarcinoma was already most common among patients <50 years of age by 1973-1974, while adenocarcinoma rapidly increased and surpassed squamous carcinoma in all other age groups by 1990-1994.

      Conclusion:
      Incidence of adenocarcinoma of the lung has continued to increase to such an extent that it comprises a clear majority of all lung cancers in the US. Indeed, our analysis demonstrated that lung adenocarcinoma currently represents 55% of US lung cancers. It is the most common histology in men and women, in whites, blacks, and other-races, and in all age groups. The question of how the actions of Big Tobacco helped to create this epidemic will be addressed in a separate presentation at this meeting.

      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.

  • +

    P3.05 - Poster Session/ Prevention and Tobacco Control (ID 217)

    • Event: WCLC 2015
    • Type: Poster
    • Track: Prevention and Tobacco Control
    • Presentations: 1
    • +

      P3.05-001 - The Role of Big Tobacco in the Creation of the Expanding Epidemic of Smoking-Related Adenocarcinoma of the Lung (ID 2492)

      09:30 - 17:00  |  Author(s): M. Finkelman

      • Abstract
      • Slides

      Background:
      In 1950, when the relationship between cigarette smoking and lung cancer was definitively demonstrated, adenocarcinoma of the lung comprised approximately 5% of lung cancers and appeared to be unrelated to smoking. Subsequently, the incidence of lung adenocarcinoma increased sharply, and became strongly related to smoking. Utilizing SEER data on 419,941 lung cancers diagnosed between 1973 and 2011, we demonstrate that adenocarcinoma now comprises 55% of all lung cancers in the US. Adenocarcinoma rose in conjunction cigarette design changes introduced by the Tobacco Industry beginning in the 1950s in response to mounting evidence that smoking caused other forms of lung cancer. The objective of this abstract is to address how actions of Big Tobacco were primarily responsible for the rise of adenocarcinoma of the lung.

      Methods:
      Because SEER contains no information about cigarette smoking, other sources were utilized to correlate changing histology to time trends in smoking prevalence, the changing cigarette, and Tobacco Industry actions. These include internal Tobacco Industry documents, historical documents describing Tobacco Industry actions, several Surgeon General Reports, NCI Monograph #13, and the verdict of Civil Action No. 99-2496: “United States versus Phillip Morris et al.”

      Results:
      Mounting evidence from population-based epidemiological analyses, the 1953 mouse painting experiments, and extensive press reporting created a crisis for the Tobacco Industry, as smoking rates temporarily dropped in the early/mid 1950s. While the Tobacco Industry consistently denied the evidence, they introduced filters and low yield cigarettes, inferring that modifications in cigarette design were safer. Indeed, many public health professionals believed that there would be some benefit from these changes insofar as compensation by smokers would be incomplete. Moreover, numerous epidemiologic studies appeared to support that filtered and low yield cigarettes conferred a lower lung cancer risk. Indeed, the 1981 Surgeon General Report on "The Changing Cigarette" concluded that individuals unable to quit should switch to filtered and low tar cigarettes. It was not until the analysis of Brown & Williamson internal documents in 1994 and other previously secret Tobacco Industry documents after the Master Settlement Agreement in the 1998 that it became abundantly clear regarding the extent to which the Tobacco Industry had knowingly deceived both the public and federal government about the safety of cigarette design changes for decades.

      Conclusion:
      Big Tobacco intentionally and extensively deceived the public during the second half of the 20th century. Trends in the rising incidence of adenocarcinoma of the lung correlate with the wide-scale adoption by smokers of filtered and low-yield cigarettes. Actions of Big Tobacco were predominantly responsible for the current epidemic of smoking-related lung adenocarcinoma.

      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.

  • +

    P3.06 - Poster Session/ Screening and Early Detection (ID 220)

    • Event: WCLC 2015
    • Type: Poster
    • Track: Screening and Early Detection
    • Presentations: 1
    • +

      P3.06-003 - Effectiveness of Lung Cancer Screening Comparing Computed Tomography (CT) to Chest X-Ray (CXR) to No Screening in PLCO and NLST Randomized Trials (ID 725)

      09:30 - 17:00  |  Author(s): M. Finkelman

      • Abstract
      • Slides

      Background:
      We sought to estimate the relative effectiveness of CT and CXR versus no screening in the context of a comparative cost-effectiveness analysis of lung cancer screening. CXR is considered ineffective because no randomized population trial (RPT) has shown a lung cancer mortality reduction. In the Mayo Lung Project, however, CXR screening produced a significant survival advantage which was not attributable to overdiagnosis or other screening biases (JCO 20:1973-83; 2002). The lung portion of the Prostate Lung Colon Ovary (PLCO) Cancer Screening Trial reported no lung cancer mortality reduction with CXR versus no screening, and it is considered to be a negative trial. The National Lung Screening Trial (NLST) was the first lung cancer screening RPT to report a mortality reduction comparing CT to CXR, but lacked an unscreened control. Survival from these trials has not been reported.

      Methods:
      To compare effectiveness of CXR and CT versus no screening, we calculated mortality, survival, and stage distribution in an intent-to screen analysis of PLCO and NLST data. Only lung cancers diagnosed within 7 years of randomization in PLCO were considered to match the median 6.7 years follow-up in NLST. Kaplan-Meier survival was compared by the log-rank test. Incidence, mortality, and stage distribution were compared with Fisher’s exact test. All p-values are two-sided.

      Results:
      In PLCO, 154,897 participants were randomized to either four annual CXRs over 3 years or to no screening. Within 7 years of randomization, 1072 and 1022 lung cancers were diagnosed, respectively (RR=1.05; 95%CI 0.96-1.14; p=0.271). 5-year survival was 27% and 18% (p<0.001). Mortality analysis revealed 764 and 811 lung cancer deaths in CXR and control groups (RR=0.94; 95%CI 0.85-1.04, p=0.244). The CXR group had significantly more stage IA cancers (RR 1.70; 95%CI 1.33 – 2.16; p<0.001) and fewer advanced stage IIIB and IV cancers (RR=0.87; 95%CI 0.76 – 0.99; p=0.044). NLST randomized 53,452 participants to either three annual CTs or CXRs over 2 years. There were 1089 and 969 lung cancers in the CT and CXR groups respectively (RR=1.12; 95%CI 1.03-1.22; p=0.007). 5-year survival was 49% and 33% (p<0.001). Mortality comparisons revealed 449 and 528 lung cancer deaths in the CT and CXR groups (RR=0.850; 95%CI 0.751-0.964, p=0.012). There were significantly more stage IA cancers (RR 2.16; 95%CI 1.82 – 2.56; p<0.001) and fewer advanced stage IIIB and IV cancers in the CT versus CXR groups (RR=0.74; 95%CI 0.63 – 0.87; p<0.001).

      Conclusion:
      Based upon similar lung cancer incidence, improved survival, and a more favorable stage distribution (particularly a reduction in the number of advanced cancers) in PLCO, CXR screening is superior to no screening, and overdiagnosis does not account for this advantage. While CXR screening is superior to no screening, CT is more efficacious than CXR in NLST. As CT is more expensive, has a higher false positive rate, and is more likely to detect overdiagnosed cancers than CXR, CXR may still be cost-effective compared to CT. Accordingly, a cost-effectiveness analysis employing NLST and PLCO data is ongoing.

      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.