Analysis of "Differential Survival Among Patients With Cancer by COVID-19 Vaccination Status"

A recently published study led by M. Kelsey Kirkwood, MPH (ASCO), Charu Aggarwal, M.D., MPH (University of Pennsylvania) and other co-authors including experts from ASCO and affiliated institutions hypothesize that COVID-19 vaccination, particularly recent doses, improves overall survival (OS) in cancer patients infected with SARS-CoV-2. The study also hypothesized that vaccination rates would differ by demographics, social determinants of health (SDOH), and clinical characteristics. The authors found that hazard ratios for mortality were higher among unvaccinated individuals with cancer.

Study Design and Methodology

This cohort study Included 1,765 patients with cancer and documented SARS-CoV-2 infection (April 2021–July 2023). Note that the authors categorized participants by vaccination status into three groups:

• No dose
• Last dose 1-6 months before infection
• Last dose >6 months before infection

As far as study data collection, the data was sourced from the ASCO COVID-19 registry, including demographic, clinical, and SDOH variables. Follow-up lasted up to 2 years, with outcomes assessed via medical records.

Upon study analysis the team employed Kaplan-Meier analysis estimated OS, as well as Hazard ratios (HRs) from Cox proportional hazards models for assessment of association between vaccination and OS, adjusting for age, cancer type, performance status, and SDOH.

Associations between patient characteristics and overall survival, adjusted (n = 1,765). Black circles are estimated HRs from a single Cox multiple regression model (i.e., reported HRs are adjusted for all other covariates included in the figure); horizontal lines are 95% CIs. HS Grad refers to area-level quartiles of the distribution in the United States by zip code of residence (with Q1 being the lowest and Q4 being the highest). ECOG PS, Eastern Cooperative Oncology Group performance status; HR, hazard ratio.

Key Findings

First, the authors looked at survival outcomes. For patients vaccinated within 1-6 months before infection: 6-month OS was 90%. Patients unvaccinated or with last dose >6 months before infection: 6-month OS was 82%.

When applying adjusted HR for death:

• No dose: HR = 1.60 (95% CI, 1.26-2.04).
• Last dose >6 months ago: HR = 1.32 (95% CI, 1.03-1.69).

What about demographic and socioeconomic insights?

Older patients (>65 years), urban residents, and those with higher SDOH indices were more likely to be vaccinated. Patients with B-cell malignancies showed higher vaccination rates compared to other cancer types.

The team found that vaccine recency matters. The timing of the last dose was more significant than the total number of doses in predicting survival.

Limitations

As with all studies this one is no exception with limitations. TrialSite lists key limitations below:

• Observational Nature
• The study is observational, limiting causal inferences. Associations may be influenced by unmeasured confounding factors.
• Data Collection Variability
• Differences in clinical data recording across sites may introduce inconsistencies.
• Cause of Death Ambiguity
• OS was used instead of COVID-19-specific survival due to potential indirect mortality effects, which may obscure direct impacts of vaccination.
• Heterogeneous Population
• Inclusion of diverse cancer types and treatment regimens makes subgroup comparisons challenging.

Potential Biases and Assumptions

• Selection Bias
• Patients included may not fully represent the broader cancer population, particularly those without access to participating centers.
• Assumption of Equivalence in Comorbidities
• Adjustments for comorbidities may not fully account for variability in underlying health conditions.
• Vaccination Timing
• Categorization based on recency of the last dose assumes uniform efficacy across all vaccine types and regimens, which may not hold true.

Conclusion

This study highlights the protective role of COVID-19 vaccination in improving survival among cancer patients, especially when received within 6 months before infection. Findings support the need for more frequent vaccination schedules tailored to oncology patients, given their heightened vulnerability. While the study underscores the value of vaccination, further randomized trials or more controlled studies are necessary to refine vaccination strategies for this population.

Conflict of Interest

The authors disclosed professional affiliations with pharmaceutical companies, which could influence perspectives but did not report direct conflicts related to this study.

Lead Research/Investigator

• Kelsey Kirkwood, MPH
• Charu Aggarwal, MD, MPH
• Suanna S. Bruinooge, MPH
• Jen Hanley Williams, MA
• Melinda Kaltenbaugh, MBA
• Natalie Ramos, MPA
• Julie R. Gralow, MD, FACP, FASCO
• Elizabeth Garrett-Mayer, PhD

Source: JCO Oncology Advances