Adverse Sequelae of the Disease and Therapy

Infectious complications in advanced disease are in part a consequence of the hypogammaglobulinemia and the inability to mount a humoral defense against bacterial or viral agents. Herpes zoster represents a frequent viral infection in these patients, but infections with Pneumocystis carinii and Candida albicans may also occur. The early recognition of infections and the institution of appropriate therapy are critical to the long-term survival of these patients. A randomized study of intravenous immunoglobulin (400 mg/kg every 3 weeks for 1 year) in patients with CLL and hypogammaglobulinemia produced significantly fewer bacterial infections and a significant delay in onset of first infection during the study period.[19] There was, however, no effect on survival. Routine chronic administration of intravenous immunoglobulin is expensive, and the long-term benefit (>1 year) is unproven.[20,21]

Patients with CLL who required hospitalization for COVID-19 prior to the induction of vaccines fared poorly regardless of stage in two retrospective reports.[22,23] One of the studies noted a protective effect from Bruton tyrosine kinase (BTK) inhibitors (usually ibrutinib),[23] but this was not seen in the other report.[22] With enhanced testing and the advent of multiple therapeutic strategies to prevent and treat COVID-19, the case fatality rate for patients with CLL dropped from 35% in early 2020 to 11% in late 2020 and early 2021 (P < .001).[22] For patients requiring hospitalization, the case fatality rate dropped from 40% to 20% (P = .003).

Autoimmune hemolytic anemia and/or thrombocytopenia can occur in patients with any stage of CLL.[24] Initial therapy involves corticosteroids with or without alkylating agents (fludarabine can worsen the hemolytic anemia). It is often necessary to control the autoimmune destruction with corticosteroids, if possible, before administering marrow-suppressive chemotherapy because it may be difficult for a patient to successfully receive a red blood cell or platelet transfusion. Alternate therapies include high-dose immune globulin, rituximab, cyclosporine, azathioprine, splenectomy, and low-dose radiation therapy to the spleen.[3,25] Tumor lysis syndrome is an uncommon complication (presenting in 1 of 300 patients) of chemotherapy for patients with bulky disease.[26]

Second malignancies and treatment-induced acute leukemias may also occur in a small percentage of patients.[27] Transformation of CLL to diffuse large B-cell lymphoma (DLBCL, known as Richter syndrome) occurs in 2% to 10% of patients. Risk factors for transformation include unmutated IGH, TP53 or NOTCH1 pathogenic variants, CDKN2A/B loss, and a complex karyotype.[28] Up to 60% to 70% of patients develop a DLBCL clonally related to the CLL, and these patients have a significantly worse prognosis than patients with de novo DLBCL.[29-31] Patients with a clonally unrelated DLBCL have a much better prognosis, which is similar to de novo DLBCL.[29] However, there is limited availability for real-life sequencing of the immunoglobulin heavy chains in the original CLL sample to compare with the transformed sample. Characteristic molecular signatures may serve as an alternate way to assess prognosis.[32] Up to 20% to 40% of patients with clonally related Richter syndrome are disease free for more than 5 years after aggressive combination chemotherapy, typically R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone) or Pola-R-CHP (polatuzumab, rituximab, cyclophosphamide, doxorubicin, and prednisone), often followed by autologous or allogeneic stem cell transplant.[33-35] For more information, see Aggressive B-Cell Non-Hodgkin Lymphoma Treatment. In two retrospective reports, CLL with transformation to Hodgkin lymphoma had the same prognosis as de novo presentations of Hodgkin lymphoma at an equivalent age.[36,37][Level of evidence C3]

The BTK inhibitors increased the risk of bleeding requiring hospitalization (3-year risk for patients who received ibrutinib, 8.8% [95% confidence interval (CI), 6.5%–11.7%]) and atrial fibrillation (3-year incidence for patients who received ibrutinib, 22.7% [95% CI, 19.0%–26.6%]).[38] A randomized trial with a median follow-up of 41 months showed less atrial fibrillation for patients with CLL who received acalabrutinib compared with ibrutinib (9.4% vs. 16%; P = .02).[39]