Patients With Hematologic Malignancies: How Robust Is Their Immunity to SARS–CoV-2?

Get Permission

A large study from the United Kingdom has taken a deep dive into SARS–CoV-2 in the setting of cancer, yielding both concerning and encouraging findings about natural and vaccine-induced immunity. The study was presented at the Presidential Symposium during the European Society for Medical Oncology (ESMO) Congress 2021 by Scott T.C. Shepherd, MD, of The Royal Marsden Hospital NHS Foundation Trust and The Francis Crick Institute, London.1

“Whereas patients with solid cancers are comparable to noncancer controls, patients with hematologic malignancies had reduced or absent neutralizing activity against variants of concern, but they had preserved cellular responses that might contribute to protection,” Dr. Shepherd said. “Our results lend support to prioritization of all patients with hematologic malignancy for further booster vaccination, whereas patients with solid cancers may be prioritized by age.”

The CAPTURE trial was a longitudinal prospective cohort study sponsored by The Royal Marsden and NHS Foundation Trust, with laboratory work conducted at The Francis Crick Institute. The work was funded by a grant from The Royal Marsden Cancer Charity. Its aim was to study functional immune responses to both infection and vaccination in persons with cancer, integrating clinical data to explain the impact of disease- and treatment-specific factors on those responses.

The study evaluated two cohorts: 118 patients with confirmed COVID-19 infection (infection cohort) and 585 patients who received vaccination (vaccination cohort). The study tested for binding antibody responses; live virus neutralizing antibody responses to wild-type SARS–CoV-2 and the variants of concern Alpha, Beta, and Delta; and the presence of SARS–CoV-2–specific T cells.

Infection Cohort: Solid Tumors vs Hematologic Cancers

In the infection cohort, 89% of the 118 patients had solid malignancies, and 18% had hematologic cancers. The infection cohort was recruited between May 2020 and February 2021; thus, most subjects were infected with wild-type SARS–CoV-2 or the Alpha variant.

Patients were found to test positive for SARS–CoV-2 after a median follow-up of 154 days. Infection severity was considered asymptomatic in 24, mild in 52, moderate in 36, and severe in 6 patients.

“Previous infection induced robust and durable neutralizing responses in most patients with cancer,” Dr. Shepherd reported. The majority (82%) had S1-reactive antibodies (reacting to the S1 subunit of the spike protein) or neutralizing antibodies (89%). Levels of neutralizing antibodies were, however, lower against all variants of concern, most importantly, Delta and Beta, compared with wild-type SARS–CoV-2.

Although S1-reactive antibody levels waned in a proportion of patients over time, neutralizing antibodies against wild-type virus remained stable up to 329 days (median, 181 days). Additionally, most patients also had detectable spike-specific T cells, though the CD4-positive T-cell responses predominated over the CD8-positive T-cell responses (76% vs 52%). A multivariate model showed no association between neutralizing antibody titers and comorbidities, age, gender, or disease severity, but patients with hematologic malignancies had lower levels of neutralizing titers.

“When we compared patients with hematologic malignancies with those who had solid cancers, we saw significantly fewer of them had detectable T-cell responses,” revealed Dr. Shepherd. This impairment was disease- and treatment-specific, and there was evidence suggestive of compensation from T cells.

“Considering both humoral and cellular responses together in patients with hematologic malignancies, we noted a disconnect between the arms of the immune response. All patients with lymphoma—most of whom had received anti-CD20 monoclonal antibodies in the past 12 months—had undetectable humoral responses after infection, but they compensated for this through higher T-cell activation.”

Integration of clinical covariates into the multivariate model showed no significant impact of anticancer therapies or comorbidities on cellular responses to infection. The one exception was that CD4-positive T-cell activation was lower in patients treated with immune checkpoint blockade (odds ratio = 0.21). “It’s reassuring, though, that large-scale registry data sets have not shown this group to be at increased risk for severe COVID-19…,” Dr. Shepherd commented.

Large Vaccine Cohort

In addition, the CAPTURE trial evaluated immune response to vaccination in 585 patients, 76% with solid tumors and 24% with hematologic cancers.2 Previous COVID-19 infection was documented in 31% of patients.

The vaccine type was Oxford/AstraZeneca in 74% and Pfizer in 26%. Patients were tested for immune response 2 to 4 weeks after the first dose, immediately before the second dose (given 10–12 weeks later), and 2 to 4 weeks after the second dose.

Binding antibodies (anti-S1 IgG enzyme-linked immunoassay) were observed in 78% of the infection-naive patients with cancer after two doses but in just 48% after one dose. After two doses, 83% had detectable titers to wild-type SARS–CoV-2, but the percentage dropped to 61% vs Alpha, 53% vs Beta, and 54% vs Delta. This corresponded to a more than twofold reduction in median titers for variants of concern vs wild-type virus.

“This showed that measuring binding antibodies to wild-type spike, which is routine in clinical practice, may overestimate protection to variants of concern, and may be falsely reassuring to patients in terms of their degree of protection after vaccination,” Dr. Shepherd commented.

“Previous SARS–CoV-2 infection boosted the neutralizing antibody response, lending support to a third vaccine dose in patients with cancer,” he added. Responses to Delta were observed in 80% of patients with prior infection vs 53% of those without. This boosting of response occurred both in patients with solid tumors and hematologic malignancies.

In comparing infection-naive patients with solid tumors with those who have hematologic malignancies, tests showed that neutralizing activity was significantly reduced in the latter, especially to variants of concern. Among patients with hematologic malignancies, 69% had undetectable antibodies to Delta after two doses, he reported.

“Patients with chronic lymphocytic leukemia, myelodysplastic syndrome, and lymphoma had the lowest neutralizing activity to wild-type SARS–CoV-2; however, there were no significant differences in neutralizing antibody activity to Delta, where the majority of patients with blood cancer had undetectable neutralizing antibodies, regardless of the subtype,” he further explained.

The multivariate model showed that chemotherapy, targeted therapy, immunotherapy, and other immunosuppressive treatments do not appear to impact immune response. The one exception was treatment within 12 months with anti-CD20 monoclonal antibodies, “which profoundly suppressed humoral response to vaccination,” he said.

Humoral responses in the CAPTURE trial were compared with those from the Oxford/AstraZeneca Legacy cohort of healthy volunteers, which previously established that immune response declines with age.3 In a comparison of the two data sets by age, no significant differences in median antibody titers were shown between patients with solid cancers and healthy volunteers.

T-Cell Responses in Infection-Naive Patients

SARS–CoV-2–reactive T-cell responses after two doses were detected in 79% of infection-naive patients.4 “We also detected T cells reactive to Alpha and Delta peptide pools, indicating that unlike the neutralizing antibody responses, T cells respond to a broad range of epitopes that cross-react with wild-type or other variants of concern,” Dr. Shepherd explained. “Even in patients with absent neutralizing activity—such as those with anti-CD20 monoclonal antibody treatment—we were able to detect T-cell responses.”

Although median titers were slightly lower in patients with hematologic malignancies vs solid tumors, a similar proportion of patients in each group did have detectable T-cell responses. No association was shown among cancer diagnosis, age, comorbidities, or treatment in predicting T-cell response.

For more on the use of vaccination for COVID-19 in patients with solid tumors, see the findings of a large Dutch study also presented during the ESMO Congress 2021 and reported in The ASCO Post

DISCLOSURE: Dr. Shepherd reported no conflicts of interest.


1. Shepherd STC, Fendler A, Au L, et al: Adaptive immunity to SARS–CoV-2 infection and vaccination in cancer patients: The CAPTURE study. ESMO Congress 2021. Abstract 1557O. Presented September 20, 2021.

2. Fendler A, Shepherd STC, Au L, et al: Adaptive immunity and neutralizing antibodies against SARS-CoV-2 variants of concern following vaccination in patients with cancer: The CAPTURE study. Nat Cancer. October 27, 2021 (early release online)

3. Wall EC, Wu M, Harvey R, et al: AZD1222-induced neutralising antibody activity against SARS–CoV-2 Delta VOC. Lancet 398:207-209, 2021.

4. Fendler A, Au L, Shepherd STC, et al: Functional antibody and T cell immunity following SARS-CoV-2 infection, including by variants of concern, in patients with cancer: The CAPTURE study. Nat Cancer. October 27, 2021 (early release online)