Does location matter? Cancer and COVID-19

The COVID-19 pandemic quickly swept across the United States (US) in 2020, bringing with it a myriad of challenges, loss of life, and presenting a tremendous burden on healthcare facilities nationwide. As the pandemic has evolved over time and more is known about the virus and how it operates, effective treatments have been developed that help to combat the side-effects of contracting COVID-19. However, people living with cancer remain a high-risk group who are at an increased likelihood of developing severe complications. Dr. Jessica Hawley, a new faculty member in Fred Hutch’s Clinical Research Division, and colleagues, as part of the COVID-19 and Cancer Consortium, sought to understand whether geographical region plays a role in COVID-19 outcomes for patients living with cancer. They reasoned that by understanding regional differences, it would allow for better allocation of resources and efforts to improve standard of care for this population. Their study, recently published in JAMA Network Open, was formed based on experiences had while working on the frontlines early in the pandemic and observing changes in trends as time went on. “We assumed as more knowledge was gained about COVID-19 that outcomes might improve but how might resource allocation, case-volume, and being treated at an academic center (or not) factor into patient outcomes? Thus, the question arose, would patients have differential outcomes by region after controlling for seasonality and other confounders? We hypothesized that outcomes would be associated with region. Reassuringly, for patients treated across the country, we did not observe this to be the case,” explained Dr. Hawley.

The COVID-19 and Cancer Consortium is comprised of over 100 cancer centers across the US and actively collects clinical data about COVID-19 in cancer populations; clinical data which provided the basis of this study. With the study population defined and comprehensive characteristics including age, treatment, and comorbidities available, the authors next focused on geographical location outlining 9 US Census Bureau divisions, described in their results as 4 geographical regions, to stratify the patient population into. Interestingly, while investigating patient characteristics, the authors observed a difference by region in treatment modality. They surmised that this was related to divergent regions peaking in case number at different times throughout the pandemic, and the ever-changing landscape of available treatments. 

Next, they examined outcomes at a cancer center level all of which were in metropolitan areas. They discovered contrasting outcomes between cancer centers, with those based in metropolitan areas with smaller populations having reduced 30-day mortality rates (the primary study outcome) compared to those in more highly populated areas. Interrogating the dataset further, they noted that presenting with severe COVID-19 was more common in highly populated areas, a finding they hypothesized could be linked to social and environmental factors. At a regional level, the Northeast region had higher 30-day mortality levels compared to the South, Midwest and West, with utilization of mechanical ventilation slightly higher in cancer centers in the Southern region. However, importantly, there were no statistically significant differences in COVID-19 outcomes across all the regions. Further, the region lived in by a person with cancer who had COVID-19 did not increase their risk of overall mortality. Summarizing these findings Dr. Hawley stated “although we did not find outcome differences by region, we did observe notable heterogeneity, or differences, in outcomes across cancer centers […]. In other words, there was variation in outcomes depending on which cancer center patients were treated.”

As uncertainties attributed to a future with COVID-19 remain and we continue to learn and adapt, this study has highlighted existing heterogeneity in COVID-19 outcomes amongst cancer centers and is a step towards understanding the specific needs held by people with both cancer and COVID-19. Going forward, “potential next steps suggested from this study would be to explore the specific mechanisms that led to this variation in outcomes across cancer centers. Once mechanisms are identified, the clinical oncology community could be intentional about directing resources and quality improvement projects towards those inequities,” said Dr. Hawley. She continued by highlighting specific examples including how “many healthcare volunteers from across the nation came to NYC [New York City] to volunteer and assist in the short-term, filling a void for the much-needed extra personnel […]. It would be informative to know if these additional resources were available at differential levels across cancer centers and if the implementation of these efforts contributed to improved outcomes.”

This work was partly supported by grants from the National Cancer Institute.

Hawley JE, Sun T, Chism DD, Duma N, Fu JC, Gatson NTN, Mishra S, Nguyen RH, Reid SA, Serrano OK, Singh SRK, Venepalli NK, Bakouny Z, Bashir B, Bilen MA, Caimi PF, Choueiri TK, Dawsey SJ, Fecher LA, Flora DB, Friese CR, Glover MJ, Gonzalez CJ, Goyal S, Halfdanarson TR, Hershman DL, Khan H, Labaki C, Lewis MA, McKay RR, Messing I, Pennell NA, Puc M, Ravindranathan D, Rhodes TD, Rivera AV, Roller J, Schwartz GK, Shah SA, Shaya JA, Streckfuss M, Thompson MA, Wulff-Burchfield EM, Xie Z, Yu PP, Warner JL, Shah DP, French B, Hwang C; COVID-19 and Cancer Consortium (CCC19). Assessment of Regional Variability in COVID-19 Outcomes Among Patients With Cancer in the United States. JAMA Netw Open. 2022 Jan 4;5(1):e2142046. doi: 10.1001/jamanetworkopen.2021.42046. PMID: 34982158; PMCID: PMC8728628.