December 05, 2019
3 min read

Location linked to likelihood of receiving high-cost drugs for advanced lung cancer

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Cathy Bradley
Cathy J. Bradley

Patients with a primary diagnosis of metastatic non-small cell lung cancer appeared more likely to receive newer, high-cost antineoplastic agents at NCI-designated facilities compared with other settings, according to results of a retrospective observational study published in Journal of the National Cancer Institute.

Those who lived in high-poverty areas appeared less likely to receive the high-cost drugs, researchers found.

“Costs of new cancer drugs are escalating well beyond reach for many patients — even for those who are insured,” Cathy J. Bradley, PhD, associate dean for research at Colorado School of Public Health and deputy director of University of Colorado Cancer Center, told Healio. “We were interested in whether agents that cost more than $5,000 per month would widen disparities in cancer outcomes between patients who are disadvantaged either by income, residence or access to specialty care compared with those who are not.”

Bradley and colleagues analyzed SEER-Medicare data on 10,655 patients aged 66 years and older who were enrolled in fee-for-service Medicare Part D and treated with an antineoplastic agent for a first primary diagnosis of advanced-stage NSCLC between 2007 and 2015. Researchers defined high-cost agents as those costing $5,000 or more per month.

Independent variables included race/ethnicity, urban or rural residency, census tract poverty and type of treatment facility. Follow-up was 1 year after the month of cancer diagnosis.

The average overall monthly cost of drugs during the study period was $13,029 (standard deviation, $10,201). Median OS was 1 year; 60% of patients died during the observation period.

Factors associated with a greater likelihood of receiving high-cost agents included female sex, white non-Hispanic race, residence in urban areas or counties with less than 20% of the population below the poverty level, and fewer comorbid conditions.

Marginal effects of logistical models showed patients who resided in high-poverty vs. low-poverty areas appeared four percentage points less likely to be treated with high-cost agents (two-sided P < .001). Patients not treated at NCI-designated cancer centers appeared 9.5 percentage points less likely to receive high-cost agents than those treated at such centers (P < .001), whereas those who resided in an area without vs. with a medical school-affiliated hospital appeared 4.2 percentage points less likely (P = .002).

Researchers observed a 27 percentage-point increase (two-sided P < .001) from 2007 to 2015 in the likelihood of treatment with a high-cost agent, which highlights the rapid change in practice patterns, according to the researchers.


Medicare spending for patients who received high-cost agents within 1 year of diagnosis was $93,953 compared with $58,880 for those who did not receive high-cost agents (P < .001).

A smaller percentage of patients in rural vs. urban areas (60.4% vs. 67.8%) received high-cost agents. Medicare spending within 1 year of diagnosis was about $17,000 lower per patient among those residing in rural vs. urban areas.

The researchers reported several study limitations, including the exclusion of novel agents approved beyond 2015 and a lack of data on patients’ preferences, functional status and quality of life, as well as adverse events and morbidity associated with treatment. They also acknowledged that the study’s elimination of patients who did not receive antineoplastic agents resulted in the disproportionate exclusion of those who resided in rural or low-poverty areas.

“Although we did not study the appropriateness of care, the findings suggest that a research-oriented facility more readily embraces new therapies — perhaps because they are more familiar with use and how to manage side effects,” Bradley told Healio. “In addition, many of the medications prescribed require biomarker testing, which may not be available in some settings.

“We would like to examine in future research whether biomarker testing occurred and if medications were prescribed when indicated through biomarker testing, and whether these prescribing patterns differed by setting and patient characteristics,” Bradley said. “We are also interested in whether outcomes differed between patients who received the high-cost agents relative to those who did not.” – by Jennifer Southall

For more information:

Cathy J. Bradley, PhD, can be reached at University of Colorado Comprehensive Cancer Center, 13001 E. 17th Place, B119, Bldg. 500, Room N6203L, Aurora, CO 80045; email:

Disclosures: NCI funded the study. Bradley reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.