Steroids Increase Infection Risk With CAR T-Cell Therapy in NHL

Liam Davenport

December 10, 2020

Giving steroids to patients with relapsed/refractory high-grade B-cell non-Hodgkin's lymphoma (B-NHL) following chimeric antigen receptor (CAR) T-cell therapy may place them at substantially increased risk of infection, suggests a real-world analysis of UK patient data.

The analysis also indicated that patients who have undergone multiple prior lines of chemotherapy are also at particular risk after receiving the treatment.

The research was presented at the virtual 62nd ASH Annual Meeting and Exposition on December 7.

London Study

Dr Lorna Neill, Department of Haematology, University College London Hospitals NHS Foundation Trust, and colleagues looked at the records of 60 high-grade B-NHL patients given CAR-T cell therapy, more than a third of whom had received at least three prior lines of therapy.

Almost half of the patients developed an infection within 30 days of receiving treatment, the majority bacterial in nature. Over half also developed cytokine release syndrome (CRS) and a quarter severe immune effector cell-associated neurotoxicity syndrome (ICANS).

Analysis showed the risk of infection following CAR T-cell therapy was increased 4.3-fold in patients who had received at least three prior lines of therapy, while steroid use increased the risk by a factor of 3.8.

Presenting the results, Dr Neill said that the rate of early infections post-CAR T-cell infusion is "comparable with other cohorts".

However, she noted that, "unlike in some US centres, this cohort did not receive prophylactic antibiotics or intravenous immune globulin".

Dr Neill continued: "Clearly, patients with advanced disease with high-grade lymphoma are at high-risk for CRS and ICANS, which increases the use of steroids and appears to lead to infectious complications," Dr Neill added.

"Heavy pre-treatment may also increase the risks by intensifying immunosuppression prior to CAR T-cell therapy”, and "strategies to modulate these risks include optimisation of [chemotherapy] bridging to reduce the disease burden...combined with infectious prophylaxis from referral until at least 3 to 6 months post-infusion."

She said: "In this analysis, steroids represent a significant risk and efforts should be made to wean doses swiftly," while the use of steroid-sparing agents "may be important and clinical trial results are awaited".

'Real-world Issue'

Dr Catherine Diefenbach, director, Clinical Lymphoma Program at NYU Langone's Perlmutter Cancer Center, New York, USA, who was not involved in the study, told Medscape News UK that infections post-CAR T-cell therapy "is a real-world issue".

However, she said that it "perhaps wasn't as prominent when CAR T-cells were restricted to specialised centres," and is "something we're definitely going to have to watch," as CAR T-cells have "not insignificant up-front toxicity" due to ICANS and CRS.

Dr Diefenbach also noted that it seems from the study that "there are some ways to risk-stratify and potentially prophylax these patients".

More importantly, the results are "another reason to argue against prolonged doses of steroids", she said, adding that prophylactic antibiotics may potentially be of use in particularly high-risk patients.

The CD19-targeting CAR T-cell therapies tisagenlecleucel (Kymriah, Novartis) and axicabtagene ciloleucel (Yescarta, Gilead) have been approved for use on the NHS for high-grade B-NHL since December 2018.


Dr Neill said that toxicities such as CRS and ICANS are "well reported" but "it is becoming more appreciated that this group of patients also experience a significant degree of infection," although reported rates vary.

To investigate further, the team looked at all infections in patients treated with licensed CAR T-cell at their institution between May 2019 and July 2020, dividing them into those that occurred within 30 days and 30 days or later following infusion.

They defined infections based on positive microbiological or virological results plus clinical symptoms or, in the case of invasive fungal infections, on the revised EORTC criteria, grading them as mild, severe or life-threatening.

The study included 60 adults with high-grade B-NHL, who had a median age of 60 years. Thirty-five percent were female. Thirty-seven percent of patients had received at least three prior lines of chemotherapy.

Dr Neill pointed out that the vast majority (90%) of the patients required bridging therapy, over half of whom did not respond. So, "unsurprisingly, over two thirds of patients had advanced stage disease pre-lymphodepletion and a quarter...had bulky disease".

"This highlights the refractory and high-risk nature of these patients."

Within 30 days of CAR T-cell therapy infusion, 40 episodes of infection were recorded in 28 (47%) patients, while 27 episodes occurred in 13 (22%) at 30 days or beyond.

Severe infections were observed in 15% of patients, while 12% had life-threatening infections, while in 28% of cases the infection was mild.

In the majority of cases, the infections were bacterial or respiratory viral, with bacterial infections more likely within 30 days of treatment infusion and respiratory viral infections more commonly seen later.

Dr Neill said that CRS was "common", with 52% of patients experiencing grade 2/3 CRS, and 57% required one dose of tocilizumab, but "few requiring steroids", at 8%.

In addition, ICANS grade 3–5 was reported in 25% of patients, with 28% requiring steroids for any grade of ICANS.

Bacterial infections were "clustered around steroid treatment," Dr Neill added, as were early viral reactivations.

Multivariate analysis showed that risk factors significantly associated with any infection post-CAR T-cell infusion were stage ≥3 pre-lymphodepletion, at an odds ratio of 4.2 (p=0.023) and steroid use, at an odds ratio of 3.3 (p=0.049).

Risk factors significantly associated with infection occurring less than 30 days post-infusion were at least three prior lines of therapy, at an odds ratio of 4.3 (p=0.021), and steroid use, at an odds ratio of 3.8 (p=0.03).

There was no association between CRS and the risk of infection, nor with the use of tocilizumab.

No funding declared.

Neill declares Novartis: Other: Funded attendance at academic conferences; Celgene: Other: Funded attendance at academic conferences.

62nd ASH Annual Meeting and Exposition: Abstract 3274. Presented 7 December.

Blood 2020; 136 (Supplement 1): 20–21. doi: 10.1182/blood-2020-138865


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: