Background: Cancer-directed therapy, including cytotoxic chemotherapy and immunotherapy, is a known risk factor for decreased SARS-CoV-2 seroconversion after vaccination, as a result of impaired viral immunity. Outcomes of COVID-19 infection are worse in patients on active treatment, especially those with hematologic malignancies or undergoing stem cell transplantation. The association between cancer subtypes and seroconversion rates is not well established. The goal of this study was to investigate predictive factors for SARS CoV-2 seroconversion and infection. Methods: We reviewed the electronic medical records of all patients with documented SARS CoV-2 antibody levels (between 7/2020-8/2021) in one of the largest integrated health systems in Pennsylvania. Patients who had a diagnosis of hematologic or solid malignancy, were on active treatment (including adjuvant), and had received ≥1 dose of an FDA authorized SARS CoV-2 vaccine were included. Data regarding SARS CoV-2 serology, vaccination, cancer history, and treatment regimens for each patient were collected systematically. Positive serology (reflecting seroconversion) was defined as any value ≥0.8 units/mL. Logistic regression analyses were used to examine predictors of seroconversion. Data was analyzed using SPSS v26 (IBM Corp). Results: A total of 292 patients met the inclusion criteria. Hematologic malignancy was present in 80.5% of the patients and solid tumors in 26.7%. Active disease was present in 71.6% of the cohort. Two vaccine doses were given to 92.5% of the patients and 54.8% of patients received a booster dose (95.2% received mRNA vaccines, either BNT162b2 (Pfizer BioNTech) or mRNA-1273 (Moderna)). A history of COVID-19 infection was present in 15.1% of patients, 59.1% of whom were seropositive. Rates of seroconversion were equivalent in those who received the BNT162b2 (70.2%) or mRNA-1273 (70.9%) vaccines. Seroconversion rates were 69.9% in those with active disease, 81.9% without evidence of disease, and varied by diagnosis as follows (indolent B-cell lymphoma 73.3%, aggressive B-cell lymphoma (BCL) 55.9%, plasma cell dyscrasia 80.1%, CLL 39.5%, myeloid disorder 90.2%, lung cancer 80%, breast cancer 80%, GI cancers 81.3%, and GU cancers 76.9%). On univariate regression, receipt of treatment for CLL (OR 5.79, 95% CI 2.92-11.48, p < 0.001) and aggressive BCL (OR 2.44, 95% CI 1.17-5.09) were predictive of negative serology. Results were not changed on multivariate regression when adjusted for age and active treatment. Conclusions: In this retrospective cohort of vaccinated patients on cancer-directed therapy, treatment for aggressive BCL and CLL was associated with negative seroconversion. In addition, more breakthrough infections occurred in seropositive patients, suggestive of underlying immunodeficiency related to treatment or decreased vaccine efficacy despite formation of virus-specific antibodies.