Sobuzoxane (MST-16), etoposide, and rituximab (Rituxan; R-PVP) used in combination were found to prolong survival and showcase a tolerable safety profile in patients with previously untreated diffuse large B-cell lymphoma (DLBCL) who were aged 80 years and older, according to findings from a retrospective analysis presented at the 2023 ESMO Congress.1
At a median follow-up of 35.7 months (range, 9.3-93.8), the median progression-free survival (PFS) was 17.2 months (95% CI, 7.4-44.4), the 2-year PFS rate was 44% (95% CI, 24%-62%), and the 5-year PFS rate was 22.6% (95% CI, 6.6%-44.3%). The median overall survival (OS) was 32.0 months (95% CI, 16.2-not reached [NR]), the 2-year OS rate was 62% (95% CI, 42%-77%), and the 5-year OS rate was 31.7% (95% CI, 13.4%-51.8%).
“Very old patients usually have various comorbidities, low performance status, and poor frailty,” Kaname Miyashita, MD, lead study author, of the Department of Hematology at NHO Kyushu Cancer Center in Fukuoka, Japan, and colleagues, wrote in a poster of the data. “Very aged patients [often have difficulties undergoing] R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone] therapy, which is a standard, curative chemotherapy for patients with lymphomas.”
This retrospective study evaluated the efficacy and safety of the attenuated chemotherapy regimen R-PVP in 30 patients at least 80 years of age with newly diagnosed DLBCL who received treatment in the Department of Hematology at the NHO Kyushu Cancer Center between 2008 and 2022.
Patients received an initial 4 doses per cycle of oral sobuzoxane at 400 mg and oral etoposide at 25 mg every other day on odd days plus intravenous rituximab at 375 mg/m2 on day 1 of each cycle. The number of oral doses of sobuzoxane and etoposide was increased or decreased based on the patient’s bone marrow function. Patients could receive granulocyte colony–stimulating factor at the physician’s discretion. Patients received R-PVP in 4-week intervals for a maximum of 8 cycles unless unacceptable disease control or adverse effects (AEs) occurred. The median numbers of treatment cycles of rituximab and sobuzoxane/etoposide (PVP) were both 6 (range, 1-8). Patients received a median of 4 doses (range, 1-7) of PVP per cycle.
Patients (n = 30) had a median age of 84 years (range, 80-93), 15 patients each were male and female, and the median Charleston Comorbidity Index (CCI) score was 4 (range, 2-11). Regarding DLBCL frailty, 13%, 40%, and 47% of patients were fit, unfit, and frail, respectively. Additionally, 23%, 53%, 17%, and 7% of patients had an ECOG performance status of 0, 1, 2, and 3, respectively. Most patients had non–germinal center (GC) disease (64%) and CD5-negative disease (83%). In total, 10%, 10%, and 80% of patients had stage II, III, and IV disease, respectively, and 17%, 43%, and 40% of patients had 0, 1, and at least 2 extranodal lesions, respectively. Most patients (97%) did not have B symptoms. The median lactate dehydrogenase (LDH) level was 238 IU/L (range, 135-1528), the median β2MG level was 3.1 mg/L (range, 1.5-9.3), the median sIL-2R level was 1230 U/mL (range, 299-15186), and the median albumin level was 3.7 g/dL (range, 2.1-4.6). Furthermore, 3%, 23%, 40%, and 33% of patients had low-, low-intermediate–, high-intermediate–, and high-risk disease, respectively, per the International Prognostic Index (IPI).
The overall response rate was 57%, which included a complete response (CR) rate of 40%. CRs, partial responses, stable disease, and progressive disease were observed in 12, 5, 3, and 10 patients, respectively.
The univariate hazard ratios (HRs) for PFS by subgroup were 0.70 (95% CI, 0.26-1.86; P = .473) for patients aged 85 years or older, 1.39 (95% CI, 0.55-3.53; P = .490) for males, 1.18 (95% CI, 0.37-3.79; P = .784) for those with a CCI score of at least 5, 1.50 (95% CI, 0.56-4.04; P = .419) for frail patients, 1.07 (95% CI, 0.40-2.88; P = .897) for those with non-GC disease, 2.31 (95% CI, 0.72-7.41; P = .159) for those with CD5-positive disease, 16.69 (95% CI, 4.35-64.09; P < .0001) for those with LDH levels over 1.5 times the upper limit of normal (ULN), 10.26 (95% CI, 2.14-49.10; P = .0004) for those with β2MG levels of at least 3.0 mg/L, 3.18 (95% CI, 1.19-8.51; P = .021) for those with sIL-2R levels of at least 2000 U/mL, 0.78 (95% CI, 0.27-2.21; P = .637) for those with albumin levels of at least 3.5 g/dL, and 8.40 (95% CI, 1.11-63.36; P = .039) for patients with IPI intermediate-high– or high-risk disease.
Per a multivariate analysis, the respective HRs for PFS for patients with LDH levels over 1.5 times the ULN and patients with β2MG levels of at least 3.0 mg/L were 11.80 (95% CI, 2.66-52.40; P = .001) and 8.77 (95% CI, 1.74-44.28; P = .009).
The univariate HRs for OS by subgroup were 0.87 (95% CI, 0.33-2.27; P = .774) for patients aged 85 years or older, 1.50 (95% CI, 0.57-3.94; P = .410) for males, 1.80 (95% CI, 0.65-5.01; P = .258) for those with a CCI score of at least 5, 1.57 (95% CI, 0.60-4.15; P = .359) for frail patients, 1.19 (95% CI, 0.42-3.34; P = .740) for those with non-GC disease, 2.64 (95% CI, 0.80-8.67; P = .110) for those with CD5-positive disease, 8.16 (95% CI, 2.58-25.87; P = .0004) for those with LDH levels over 1.5 times the ULN, 3.58 (95% CI, 1.18-10.86 P = .024) for those with β2MG levels of at least 3.0 mg/L, 2.83 (95% CI, 1.06-7.54; P = .037) for those with sIL-2R levels of at least 2000 U/mL, 1.20 (95% CI, 0.45-3.19; P = .709) for those with albumin levels of at least 3.5 g/dL, and 1.86 (95% CI, 0.53-6.50; P = .333) for patients with IPI intermediate-high– or high-risk disease.
Per a multivariate analysis, the HR for OS for patients with LDH levels over 1.5 times the ULN was 8.16 (95% CI, 2.58-25.87; P = .0004).
Patients with LDH levels at or below 1.5 times the ULN had a median OS of 49.8 months (95% CI, 31.1-NR), and those with LDH levels over 1.5 times the ULN experienced a median OS of 7.1 months (95% CI, 2.2-31.9; log-rank P = .002).
Regarding safety, grade 3 neutropenia was observed in 23% of patients, and grade 4 neutropenia was observed in 30% of patients. The most common non-hematologic grade 3 AEs included febrile neutropenia, which occurred in 10% of patients; infection of unknown origin and constipation, which occurred in 7% of patients each; and pneumonia, urinary tract infection, gastric ulcer bleeding, lower gastrointestinal bleeding, hematuria, hyperkalemia, and hypertension, which each occurred in 3% of patients. No grade 4 non-hematologic AEs were observed.
The investigators subsequently analyzed 45 patients with DLBCL who received either R-PVP (n = 20) or R-miniCHOP (rituximab and reduced-dose CHOP; n = 25) at the NHO Kyushu Cancer Center over the past decade. In this cohort, the median CCI score was 4 (range, 2-11) in the R-PVP arm and 2 (range, 2-5) in the R-miniCHOP arm (P = .017).
This cohort had no statistical PFS or OS differences between the patient arms per the Kaplan-Meier method. The median PFS was 18.4 months (95% CI, 7.4-NR) in the R-PVP arm vs 26.4 months (95% CI, 14.3-NR) in the R-miniCHOP arm (log-rank P = .622). The median OS was NR months (95% CI, 8.3-NR) in the R-PVP arm vs 89.1 months (95% CI, 42.2-NR) in the R-miniCHOP arm (log-rank P = .692).
Regarding safety, in the R-PVP arm, grade 3 neutropenia was observed in 30% of patients, and grade 4 neutropenia was observed in 25% of patients. The most common non-hematologic grade 3 AEs included infection of unknown origin, which occurred in 10% of patients; and febrile neutropenia, pneumonia, urinary tract infection, constipation, gastric ulcer bleeding, hyperkalemia, and hypertension, which each occurred in 5% of patients. No grade 4 non-hematologic AEs were observed.
In the R-miniCHOP arm, the most common hematologic AEs were neutropenia (grade 3, 20%; grade 4, 52%), thrombocytopenia (12%; 4%), and anemia (4%; 0%). The most common non-hematologic grade 3 AEs included febrile neutropenia, which occurred in 24% of patients; malaise, which occurred in 24% of patients; vasovagal reaction, which occurred in 16% of patients; constipation, which occurred in 12% of patients; diarrhea, which occurred in 12% of patients; dizziness, which occurred in 8% of patients; and infection of unknown origin, soft tissue infection, pneumonia, herpes zoster, influenza virus infection, acute subdural hematoma, delirium, muscle weakness of the legs, hypotension, elevated serum aspartate aminotransferase levels, and elevated serum alanine aminotransferase levels, which each occurred in 4% of patients. Additionally, 4% of patients each experienced grade 4 bronchopulmonary hemorrhage and colonic perforation.
Based on these findings, as well as findings from a phase 2 trial (NCT01087424) that investigated R-miniCHOP in patients with DLBCL aged over 80 years, the investigators proposed that very old patients with newly diagnosed DLBCL should receive supportive/palliative treatment with either R-PVP if they have LDH levels at or below 1.5 times the ULN and a high CCI score, or R-miniCHOP if they have albumin levels over 3.5 g/dL and high LDH levels.
“Our findings may suggest that R-PVP therapy is one of the safe and effective treatments for a subgroup of patients, i.e., those who may have high CCI [scores] or low tumor burden, among untreated DLBCL patients aged over 80 [years],” the investigators concluded. “Our results should be verified in a larger cohort in order to confirm and generalize the efficacy of R-PVP therapy.”
Editor’s Note: The study authors declare no competing financial interests.