7554 Background: Management of elderly DLBCL is limited by age associated comorbid conditions and decreased tolerance of chemotherapy regimens. Many elderly or unfit/frail patients (pts) with DLBCL receive either non-anthracycline based treatment (tx) or attenuated regimens such as Rituximab (R)-miniCHOP which is ~50% dose reduction from standard R-CHOP-21. Split-dose (SD) R-CHOP is an alternative approach where pts receive CHOP at a 50% dose reduction on Days 1 and 15 of each 28-day cycle with full dose Rituximab on Day 1 for 6 cycles. The cumulative dose of SD R-CHOP delivered over each 28 day cycle is equivalent in dose intensity to one cycle of standard R-CHOP-21. Methods: This is a Phase II, prospective, open label study conducted through the Wisconsin Oncology Network and included two academic centers (NCT03943901). Older pts aged≥75 years or 70-74 years plus elevated cumulative illness rating scale for geriatrics score with untreated de-novo or transformed Stage IIX-IV DLBCL were eligible. Treatment consisted of two months (4 doses) of SD R-CHOP with growth factor support followed by interim response assessment with PET/CT and peripheral blood cell-free clonoSEQ minimal residual testing (MRD). Pts with interim PET/CT score of Deauville 1-3 and MRD negativity were offered abbreviated therapy with two additional months of SD-R-CHOP. All others completed 6 cycles. The primary endpoint was end of tx (EOT) complete response (CR) rate. We utilized a Simon 2-stage design with interim analysis requiring at least 8 EOT CR in the first 16 pts (Stage I) to proceed to stage II. Results: To date, 14 pts have completed tx. The median age was 81 years (73-88) and 57% (n = 8) were male. Most pts had stage 4 disease (n = 8, 57%). 8 of 14 pts were interim PET/CT negative and 8 of 12 MRD evaluable pts were interim MRD negative. 10/14 (71%) pts achieved EOT CR exceeding our minimal threshold to continue the study early. The remaining 4 pts had progressive disease (PD) either during tx or at EOT evaluation. 6 pts were both interim MRD and PET negative; 5 de-escalated to the abbreviated arm with all achieving CR with no relapses to date. 7 pts (50%) had Grade 3+ non-hematologic toxicities and 4 pts (29%) had Grade 3-4 neutropenia. There were 2 deaths, one from COVID-19 with concurrent PD and the other from PD. Median PFS or OS have not been reached with a median follow-up of 13 months for surviving pts. Conclusions: SD-R-CHOP offers older pts with DLBCL equivalent dose intensity as R-CHOP-21 through a fractionated dosing schedule improving tolerability. Prespecified interim efficacy endpoint was met early with 10/14 pts with EOT CR. Interim stratification by MRD and PET/CT allowed de-escalation of tx in early responding pts without compromising efficacy. Clinical trial information: NCT03943901 . [Table: see text]