Infection Control in Patients With CLL May Be Pivotal for Survival Outcomes

During a presentation at the 2020 Society of Hematologic Oncology, Carsten Utoft Niemann, MD, PhD, explained that patients with chronic lymphocytic leukemia (CLL) have an increased risk for infection regardless of whether they are being treated.

With the help of the machine learning–based algorithm CLL Treatment-Infection Model (CLL-TIM), Niemann and his colleagues hope to learn whether the natural history of immune dysfunction and infections in CLL can be changed.

Investigators have made great strides in improving survival for patients with CLL, following the advent of chemoimmunotherapy and targeted therapy approaches. Unfortunately, mortality due to infection in these patients has not changed in 4 decades, Niemann explained, and infections are now the leading cause of death among patients with CLL.

“Looking at patients diagnosed with CLL, [they] tend to have a higher risk of having a serious infection within the first 5 years,” said Niemann, an associate professor, head of the CLL Lab, and chair of the Nordic CLL Study Group, in the Department of Hematology, Rigshospitalet, Copenhagen, Denmark. “The 30-day mortality [rate] upon a serious infection for patient diagnosed with CLL is 10%; the 1-year mortality [rate] is only 11% for patients starting treatment for CLL even though 50% of them will have a serious infection after starting treatment. This means that infections are actually more dangerous for patients with CLL who are untreated.”

Niemann is a coinvestigator in the phase 2 PreVent-ACaLL trial (NCT03868722), in which investigators will use the CLL-TIM algorithm to determine whether short-term treatment with the BTK inhibitor acalabrutinib (Calquence) plus the BCL-2 inhibitor venetoclax (Venclexta) can improve immune function and reduce the risk for infection compared with observation in patients with high-risk, newly diagnosed CLL.

“The aim here by this short 3-month treatment is not to cure CLL,” he said. “The aim is to reduce the CLL clone back into what we consider monoclonal B lymphocytosis and, at the same time, treat the natural history of CLL and immune dysfunction. Thus, we aim to turn back the CLL into a normal immune function, and the outcome here is actually infection-free, treatment-free survival.”

Twenty-five patients will be randomly assigned to the treatment arm, which will be given in 3 cycles of 28 days, and 25 will be assigned to observation. Enrollment has opened in Denmark and is scheduled to open soon in the Netherlands and Sweden. The primary end point is grade 3 or higher infection-free survival in the treatment arm compared with the observation arm after 24 weeks.2

CLL-TIM identifies patients at risk for infection or CLL treatment within 2 years of diagnosis as validated on both internal and external cohorts. The ensemble algorithm incorporates 28 machine learning algorithms based on data collected from 4149 patients with CLL. The model is capable of dealing with heterogeneous data, including high rates of missing data to be expected in the real-world setting, with a precision rate of 72% and a recall rate of 75%.3

Niemann added that 6 European nations are collaborating in the CLL-CLUE consortium to assess immune dysfunction and drug sensitivity in primary samples collected from patients and clinical trials, and biobanks. Long term, the goal is to better individualize CLL treatment.

Immune Disfunction Starts Long Before Diagnosis

Niemann said that patients with CLL are more likely to be prescribed macro lights, antibiotics, and/or antivirals as much as 20 years before diagnosis. “This indicates that the immune dysfunction, at least, happens decades prior to diagnosis of CLL,” he added.

Going beyond the patient’s CLL and looking at their entire medical history, Niemann said physicians can see the burden of comorbidity in this population. “Almost all types of comorbidities for patients with CLL will increase the overall mortality and also for most of these comorbidities, will increase the CLL-related mortality,” he said.

That said, it appears both treatment and CLL itself play a role in incidence of infection. Niemann noted that lower respiratory tract infections and urinary tract infections are common for the 6 months of treatment with ibrutinib (Imbruvica). “This may indicate that we have an improvement in immune function and lower degree of infections after the first half year on targeted therapy.”

Niemann pointed out that, outside of clinical trials, close to 50% of patients discontinue ibrutinib by 30 months, many due to infections. Physicians must consider that infections might also affect the continuous treatment of patients, he added.

“On the other hand, [there is] no doubt that we may, for patients who are treated, actually improve the immune function over time,” Niemann said. Patients with less disease burden in terms of being treatment-naïve have a lower risk of infections than patients who are relapsed/refractory and would be expected to have had more aggressive or longer duration of CLL.”

Cumulative mean number of infections are lower in patients whose immune globulin levels increase during ibrutinib treatment. However, Niemann said there is no clear correlation between immunoglobulin deficiency and the risk for infections.

“That is what we’ve been working on predicting the risk of infections or risk of needing treatment for CLL,” he said.

To help the patient in the clinic, Niemann said all patients with CLL may have increased infection risk, but investigators need to identify those who at greatest risk for serious infections. Further, patients with CLL are at risk for both increased frequency of infections and also in the distribution of infections, especially for pneumococcal disease.

“Vaccine responses in CLL are impaired due to immune dysfunction. Thus, it should be encouraged to have early vaccination of patients with CLL, and repeat vaccination during treatment and after treatment,” he said. “Immunoglobulin substitution should only be used for patients who both have low IgG [immunoglobulin G] levels and frequent infections as we cannot clearly define these patients just by measuring the IgG levels.”

_References

_{1. Niemann CU. Management of infections in patients with CLL. Presented at: the 2021 SOHO Annual Meeting. Virtual. September 8-13, 2021.}

_{2. Da Cunha-Bang C, Agius R, Kater AP, et al.PreVent-ACaLL short-term combined acalabrutinib and venetoclax treatment of newly diagnosed patients with CLL at high risk of infection and/or early treatment, who do not fulfil IWCLL treatment criteria for treatment. A randomized study with extensive immune phenotyping. Blood. 2019;134(supplement_1):4304. doi:10.1182/blood-2019-121907}

_{3. Agius R, Brieghel C, Andersen MA, et al. Machine learning can identify newly diagnosed patients with CLL at high risk of infection. Nat Commun. 2020;11(1):363. doi:10.1038/s41467-019-14225-8}