Purpose. To report on the risk of interstitial pneumonia (IP) in newly diagnosed diffuse large B-cell lymphoma (DLBCL) treated with dose-dense CHOP-14 plus rituximab supported with pegfilgrastim.

Patients and Methods. In this phase II study of feasibility and toxicity, 50 patients with DLBCL, aged 18–70, with no major co-morbidities were treated with standard-dose CHOP every 14 days, preceded on day 1 by rituximab at the dose of 375 mg/m2 (R-CHOP-14) and followed on day 3 by pegfilgrastim (6 mg per cycle). No cotrimoxazole prophylaxis had been devised. In patients developing symptoms of possible pulmonary complications, standard chest X-ray, high-resolution CT scan and broncho-alveolar lavage (BAL) were performed, together with complete blood count, serum immunoglobulins level and CD4 lymphocytes determination. Cytomegalovirus (CMV) infection was investigated, as well, by either CMV-antigen or CMV-IgM determination.

Results. Overall, 7 patients (14%) developed IP within 120 days from the start of therapy; their median age was 60 years (range: 22–65). Four patients developed IP after the end of therapy and no patient was neutropenic at the onset of pulmonary symptoms; the median absolute neutrophil count was 5.7x109/l (range: 3.0–11.1). Moderate or severe dyspnea and fever were constant. Diagnosis was made in all patients through high-resolution chest CT scan which documented bilateral interstitial damage, while standard chest X-ray was apparently normal in 5 of 7 cases. Clinical conditions did allow for BAL in 5 patients; Pneumocystis carinii (PC) infection was microbiologically documented in three patients, while hemorragic bronchiolitis and lymphocytic diffuse pneumonia were diagnosed in one case, each. CMV infection tests were invariably negative. Patients developing PC infection had low serum gammaglobulin levels (390, 790, and 200 mg/dL) and a lower than 400/mL number of circulating CD4+ lymphocytes (168, 359 and 190/mL, respectively). Wide spectrum antibiotics were administered in all cases until documentation of the etiology; PC infection was then treated with cotrimoxazole and low-dose steroids and lymphocytic pneumonia with high-dose steroids. The outcome was favorable in all patients, with no sequelae.

Conclusions. Our data indicate a significant risk of IP in patients with DLBCL treated with R-CHOP-14 and supported with pegfigrastim; most worrisome is the risk of PC infection in the absence of cotrimoxazole prophylaxis. The risk of IP does not correlate with neutropenia, but it correlates with a low level of gammaglobulins and of circulating CD4+ lymphocytes. In this setting of patients, cotrimoxazole prophylaxis is mandatory and high-resolution CT scan and BAL should be done as early as possible at the onset of respiratory symptoms.

Author notes

Corresponding author

Sign in via your Institution