Benjamin Levy

Cancer trends among the extreme elderly in the era of cancer screening

BACKGROUND The extreme elderly (EE; >84 years) are among the fastest growing segments of the population and bear a substantial cancer burden. We examined cancer incidence and cancer specific mortality changes among the EE during the implementation of cancer screening from the 1980s to 2000s. METHODS We examined incidence and mortality rates for breast, colon, prostate, and lung cancer by age group between 1973 and 2009 in the SEER database. We compared incidence/mortality between EE and middle aged (MA; age 50-69) patients. RESULTS Prostate cancer incidence and mortality rose and then, in the early 1990s, declined (-3.61%/year and -2.91%/year, respectively) among EE. Prostate cancer incidence rose steadily throughout the study period for MA. Breast cancer incidence rose and then declined for both MA and EE, with the decline starting in 1990 for EE (-1.34%/year), and 1998 for MA (-1.24%/year). Both age groups experienced an increase and then decrease in colon cancer incidence. The decrease in colon cancer mortality over the last decade was profound for all patients (-2.88%/year MA, and -3.29%/year EE). Lung cancer incidence (+2.35%/year to 2005) and mortality (+1.25%/year from 1995) increased for EE. Lung cancer incidence and mortality increased and then decreased (-2.54%/year for mortality from 1990) for MA. CONCLUSION Recent trends in incidence and mortality for screened cancers (breast, colon, prostate) show substantial gains for the extreme elderly, likely due in part to the effect of screening. Incidence and mortality from lung cancer, with no recommended screening during the study period, have continued to worsen for the extreme elderly, despite improvements in younger patient populations.

Circulating Tumor DNA in Non–Small-Cell Lung Cancer: A Primer for the Clinician

Circulating tumor DNA (ctDNA) consists of short, double-stranded DNA fragments that are released into the circulation by tumor cells. With the advent of newer molecular platforms, ctDNA can be detected with high sensitivity and specificity in plasma. The assay’s noninvasive nature, ability to reflect intratumoral heterogeneity, short turnaround time, and ability to obtain serial samples make it an attractive option compared with traditional tissue biopsy tumor sequencing. Currently, this technology is mostly being used for the detection of EGFR mutations in patients with advanced non–small-cell lung cancer where tissue is inadequate to detect EGFR mutations that drive acquired resistance, most notably EGFR T790M. Emerging uses include the incorporation of ctDNA testing into primary diagnosis, treatment monitoring, detection of minimal residual disease, and detection of early-stage disease in screening populations. This review summarizes both validated and evolving uses of ctDNA testing in non–small-cell l...

Influence of Extent of Lymph Node Evaluation on Survival for Pathologically Lymph Node Negative Non-small Cell Lung Cancer.

Objectives: Despite previous retrospective reports that the number of lymph nodes resected at curative intent surgery for lung cancer correlates with overall survival (OS), no consensus exists regarding the minimal nor optimal number of lymph nodes to resect at curative lung cancer surgery. Methods: We studied subjects in the Surveillance Epidemiology and End Results Database (SEER) diagnosed with non–small cell lung cancer between 2000 and 2011 who underwent either lobectomy or pneumonectomy and had pathologic negative nodal evaluation. We excluded patients with sublobar resection and/or no lymph node evaluation. We examined associations between number of lymph nodes evaluated and OS/lung cancer-specific survival by multivariable Cox regression; and predictors of evaluation of more lymph nodes. Results: Among the 33,463 patients in our sample, a median of 7 lymph nodes were evaluated. We found that lung cancer-specific survival and OS improved with increasing lymph node evaluation up to 16 to 18 lymph nodes (hazard ratio, 0.77 [95% confidence interval, 0.70-0.85] and 0.78 [95% confidence interval, 0.72-0.86], respectively). There was little additional improvement in outcomes with evaluation of >16 to 18 lymph nodes. Blacks, Hispanics, females, and patients from distinct geographical regions were less likely to have 16 or more lymph nodes evaluated. Conclusions: There was a consistently increasing survival benefit associated with a more extensive lymph node evaluation at lung cancer resection, up to 16 to 18 lymph nodes removed. The median number of nodes evaluated was, however, only 7, suggesting that setting a goal of ≥16 examined lymph nodes may lead to improved survival outcomes, and reduce disparities in care.

Physician decisions related to maintenance therapy for advanced non-small cell lung cancer (NSCLC) in the United States (US).

182 Background: In 2006, maintenance therapy was added as a treatment guideline for patients with advanced non-squamous (nsq) NSCLC. This study aimed to identify patient- and disease-related factors that impact maintenance therapy decision making. METHODS An online discrete choice survey was administered to physicians currently managing patients with advanced nsq NSCLC in the US. Physicians viewed 12 patient profiles differing in levels of the following attributes: 1st line treatment response (complete, partial, stable, progression), adverse events during 1st line therapy (none, mild, moderate, severe), comorbidities (none, mild renal, severe renal, other), patient motivation/convenience (+/+, +/-, -/+, -/-), patient insurance co-pay rate (0, 5, 10, 20%), and age (45, 58, 68, 80 years). No specific maintenance treatments were stated. For each profile, physicians indicated if they would recommend maintenance therapy. Recommendations were analyzed using a mixed-effects logistic regression model. Odds ratios (ORs) were calculated to estimate the relative odds of a maintenance therapy recommendation for levels of each attribute. RESULTS The survey was completed by 100 physicians (81% male; mean years of experience: 15.4). The study design was balanced and orthogonal. Maintenance therapy was recommended for 75% of the profiles; 98% of physicians recommended maintenance therapy for at least 1 profile, with 26% recommending it for all profiles. The odds of recommending maintenance therapy were significantly lower when the patient profile included disease progression relative to stable response (OR: 0.17; p<0.01), severe renal impairment relative to no comorbidities (OR: 0.38; p<0.01), or low motivation/convenience relative to high motivation/convenience (OR: 0.25; p<0.01). The odds also decreased with increasing age (OR: 0.97 per year increase in age; p<0.01). CONCLUSIONS Treatment response, age, motivation/convenience, and comorbidities were relevant factors for physicians when recommending maintenance therapy. Physicians report recommending maintenance even in the presence of less desirable patient and disease characteristics indicative of a real world setting.

Can racial and financial disparities be overcome in the surgical treatment of NSCLC

208 Background: Considerable data exists examining disparities in the treatment of non-small cell lung cancer (NSCLC) patients. Black patients, in particular those of lower socioeconomic status (SES), are less likely to receive appropriate care, including induction therapy and resection of surgically treatable lesions. We analyzed the outcomes of resection of NSCLC among a racially and financially diverse patient population at a large urban hospital network with a comprehensive thoracic oncology program. In this system, a patient navigation support team helped overcome barriers to preoperative preparation and multidisciplinary referral. METHODS A retrospective review of 345 patients who underwent lobectomy at our institution from 2002 - 2011 was performed. Data was retrieved from the Society of Thoracic Surgeons (STS) database and patient charts. Patient demographics, payor information and preoperative characteristics were noted. Postoperative complications, 30-day survival and 3-year survival were compared. Statistical analysis was performed using SPSS 17.0 (SPSS Inc, Chicago, IL). Chi-square test was used to compare categorical variables and Students t-test was used to compare continuous variables. RESULTS Demographics of black and non-black patients were similar. There were more black patients within the Medicaid group than non-Medicaid (48.9% and 25.3%, p=0.001). Physiologic characteristics, risk factors and use of pre-operative RT and chemotherapy were similar. Post-operative complications were comparable in Medicaid vs. non-Medicaid (11.1% and 14.7%, p=0.524), however black patients had a lower rate of complications vs. non-black (6.1% and 17.4%, p=0.007). 3-year survival was similar in the black vs. non-black (82.3% and 78.6%, p=0.879) and Medicaid vs. non-Medicaid (66.7% and 78.8%, p=0.342) groups. CONCLUSIONS We demonstrated equivalent surgical outcomes for NSCLC in addition to the similar use of induction therapy. Surprisingly, complications were lower in the black cohort. Our results reveal that appropriate treatment is being provided regardless of race or SES, and postulate that our system of preoperative patient support eliminates potential barriers to care.