(191) Soft Tissue Sarcomata in Humans at Potential Injection Sites, United States, 1975 – 2018

Publication Number: 522

Background: Vaccines are an efficient public health technology when broadly accepted, which depends on confidence in their basic safety. The current epidemic of misinformation about Covid-19 vaccines contributes to vaccine hesitancy with lethal consequences. Antipathy to immunizations, frequently associated with myths, has a long history and has repeatedly allowed vaccine preventable diseases to resurge. Pharmacoepidemiologic studies to assess new vaccine safety concerns can be time consuming and expensive. An association in cats between vaccination sites and sarcomata has been recognized for 30 years. Cases are rare but led to recommendations for immunizations at distal locations to allow resection by amputation. This risk in animals provides a basis for assessing whether a similar relationship might occur in humans.

Objectives: This proactive safety surveillance study employed U.S. cancer incidence data to seek ecologic associations between sarcoma incidence rates and potential vaccination sites.

Methods: Anticipating that feline experience might give rise to concern over possible human implications, we examined U.S. data from large cancer registries to look for ecologic associations between typical vaccine injection sites and connective tissue malignancies (excluding Kaposi’s sarcoma) between 1975 and 2018, as several new vaccines were introduced. Modeling employed SEER 9 and SEER 18 registry data covering up to 30% of the U.S. population. We defined fibrosarcoma histology by morphology codes and potential vaccine injection sites by primary tumor locations. Regression analyses employed JoinPoint software (IMS, Calverton, Maryland) to estimate average annual percentage changes in rates, change points, and the percentage change in incidence rates by year.

Results: The best fitting models from SEER 18 data have a larger decrease in fibrosarcoma incidence rates at possible vaccination sites from 2000 to 2018, with an annual percentage change (APC) of -4.5% compared to an APC of -3.3% for other locations (both significant with p < 0.001). This pattern is inconsistent with any association of these tumors with vaccinations over an extended period as vaccine coverage grew for protection against increasing numbers of target diseases. Our negative result is biologically plausible, based on established limitations in extrapolating from animals to humans along with specific demonstrations of fibrosarcomata at injection sites only in cats and ferrets.

Conclusions: We found no temporal association between fibrosarcomata in humans and potential injection sites in the U.S. between 1975 and 2018.