背景及目的：化療處方R-CHOP（內含rituximab、cyclophosphamide、doxorubicin、vincristine 及prednisolone）為瀰漫性大B 型淋巴瘤（DLBCL）的標準第一線化療。顧慮doxorubicin 可能導致心臟受損者，臨床常以處方R-CEOP 替代，而其中doxorubicin則以epirubicin 取代。在某些淋巴瘤，絕對淋巴球數值（ALC）證實可用來評估預後，然而ALC 在接受CEOP 或R-CEOP 處方的意義未明。因而本研究目的在於釐清ALC 在DLBCL 接受第一線CEOP 或R-CEOP 化療的角色。方法：回溯性分析2001 年1 月至2009 年12 月的病患特徵及治療結果。結果：總收案174 位DLBCL 病患，其中73 位接受CEOP，101 位接受R-CEOP。R-CEOP 組的無惡化存活期（PFS）顯著優於CEOP 組（p值0.0035）；此外多變項分析PFS，rituximab 証實為保護因子（危險比：0.333；95% 信賴區間：0.187~0.591；p 值< 0.001）。在治療達到完全緩解的多變項分析中，ALC 為獨立危險因子（勝算比：3.115；95% 信賴區間：1.304~7.441；p 值0.011）。在R-CEOP 組，偏低的ALC 證實預後較差（危險比：3.095；95% 信賴區間：1.335~7.180；p 值0.008），但ALC 的預測意義則不存在CEOP 組。結論： 對DLBCL 而言，rituximab 顯著延長PFS，其中ALC 偏低證實為接受R-CEOP 化療病患中較差的預後因子，但受限於回溯性研究，rituximab 及ALC 對整體存活率的影響可能低估，因而實際影響還需要長期追蹤。

Background and purpose: The R-CHOP regimen containing rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisolone is standard frontline chemotherapy for treating diffuse large B cell lymphomas (DLBCLs). Epirubicin instead of doxorubicin is administered in the R-CEOP regimen, especially for patients with concerns about cardiotoxicity. Meanwhile, the absolute lymphocyte count (ALC) shows prognostic significance in some subtypes of lymphoma. However, the role of the ALC in DLBCL patients receiving R-CEOP or CEOP is not yet clear. Hence, we attempted to elucidate the prognostic value of the ALC. Methods: Patients’ characteristics and treatment outcomes in a single institution between January 2001 and December 2009 were retrospectively analyzed. Results: In total, 174 patients with a DLBCL were enrolled. Seventy-three patients received CEOP, and 101 patients received the R-CEOP regimen. In patients receiving R-CEOP, the progression-free survival (PFS) rate was significantly higher than in those without rituximab (p = 0.0035). In addition, rituximab proved to be an independent prognosticator in the multivariate analysis of PFS (hazard ratio = 0.333; 95% confidence interval (CI), 0.187~0.591; p < 0.001). The ALC was independently predictive of complete remission in the multivariate analysis (relative risk = 3.115; 95% CI, 1.304~7.441; p = 0.011). In contrast to the CEOP counterpart, a poor prognostic value of a low ALC in PFS was demonstrated in the R-CEOP subgroup (hazard ratio = 3.095; 95% CI, 1.335~7.180; p = 0.008). Conclusions: In DLBCL, rituximab significantly increased PFS, and the ALC at diagnosis was remarkably prognostic for PFS. Because of the limitations of this retrospective study, the efficacy of rituximab and the role of ALC in the overall survival might have been underestimated. Long-term follow-up is warranted to validate these results.