化學治療引發嗜中性白血球低下之臨床表徵與顆粒性白血球生長刺激素之使用分析

Abstract

研究背景： 嗜中性白血球低下(neutropenia)與嗜中性白血球減少症併發燒(febrile neutropenia, FN)為使用化學治療之癌症病人最嚴重的併發症之一，若沒有妥善的處置，病人可能因感染而住院甚至是死亡。因此目前國外臨床治療指引皆建議，對於FN的高危險群可預防性使用顆粒性白血球生長刺激素(granulocyte colony-stimulating factor, G-CSF)。然而在臨床實務上，對於G-CSF的使用天數及高危險群的認定，仍存在許多不確定性。 研究目的： 本研究分析乳癌和非何杰金氏淋巴癌(non-Hodgkin's lymphoma, NHL)病人於接受化學治療時G-CSF之處方型態，並探討預防性使用G-CSF以及其他相關危險因子對於嗜中性白血球低下之影響。 研究方法： 本研究採回溯性世代研究，以臺灣北部某教學醫院的醫院資料庫以及病歷系統作為研究材料。研究對象為診斷為乳癌或NHL的病人，並且在2010年開始使用新的化學治療處方者。所有的病人皆以2010年開始新化學治療處方之第一天當成「進入研究世代日期(index date)」，而後追蹤觀察至化學治療或放射線療法完成，或「進入研究世代日期」後一年為止。在G-CSF處方型態分析部分，主要評估G-CSF使用目的及使用天數；在危險因子評估部分，則以邏輯式迴歸分析(logistic regression)及廣義估計方程式(generalized estimating equations, GEE)，分析預防性使用G-CSF以及其他相關危險因子對於嗜中性白血球低下的影響。 研究結果： 本研究共納入353位乳癌病人(共計2776個化學治療週期)，其中47%的病人出現嗜中性白血球低下，而其週期內發生嗜中性白血球低下的比率為13%。本研究另納入72位NHL病人(共計433個化學治療週期)，其中79%的病人出現嗜中性白血球低下，而其週期內發生嗜中性白血球低下的比率為40%。在乳癌病人的化學治療週期中，G-CSF用於初級預防的平均使用天數為4.9±2.2天；次級預防則為3.7±1.5天。相較於乳癌，NHL病人的化學治療週期中，G-CSF用於初級預防與次級預防的平均天數皆較低(初級預防：2.9±0.3天；次級預防：3.1±0.5天)。 多變項邏輯式迴歸分析中，使用G-CSF做初級預防對於乳癌病人的化學治療週期有顯著的保護效果(OR=0.49, 95%CI=0.30-0.80)，使用G-CSF做次級預防則沒有達到統計上的顯著意義(OR=0.96, 95%CI=0.67-1.37)，而其他顯著增加嗜中性白血球低下風險的因子則包括觀察期間內發生較多次嗜中性白血球低下及使用高FN風險的化學治療處方；年齡的增加則會降低嗜中性白血球低下的風險。使用GEE校正重複測量的問題後，在乳癌病人的化學治療週期中，使用高FN風險的化學治療處方為唯一增加嗜中性白血球低下風險的因子(OR=2.18, 95%CI=1.47-3.23)。 在NHL病人的化學治療週期中，多變項邏輯式迴歸分析的結果顯示，無論是使用G-CSF做初級或次級預防皆沒有顯著降低嗜中性白血球低下風險之效果(OR=5.54, 95%CI=2.79-10.99 ; OR=2.30, 95%CI=1.23-4.30)，其他顯著增加嗜中性白血球低下風險的因子有：觀察期間內發生較多次嗜中性白血球低下、使用高FN風險的化學治療處方及低血紅素(hemoglobin)。另外，GEE的分析結果則顯示，初級預防的使用對於NHL的化學治療週期並無保護效果(OR=6.49, 95%CI=2.22-19.04)，而低血紅素為增加嗜中性白血球低下風險的因子(OR=3.45, 95%CI=1.85-6.44)。邊緣區域淋巴癌(marginal zone lymphoma, MALT)相較於瀰漫性大型B細胞淋巴癌(diffuse large B-cell lymphoma, DLBCL)，則會顯著降低出現嗜中性白血球低下的風險。 研究結論： 在此教學醫院中，G-CSF預防的平均使用天數比文獻建議來的短。本研究所分析出影響乳癌及NHL病人發生嗜中性白血球低下的因子，可提供醫師做未來評估嗜中性白血球低下之高危險群的參考。

Background: Neutropenia and febrile neutropenia (FN) are major complications in cancer patients treated with chemotherapy, leading to infection-associated morbidity and mortality. Therefore, prophylactic use of granulocyte colony-stimulating factor (G-CSF) is recommended by recent clinical guidelines to manage those patients with high risk of FN. Uncertainty, however, remains in terms of the use of G-CSF, and also the clinical features of chemotherapy-induced neutropenia (CIN) in clinical settings. Objectives: The objectives of this study were to document the utilization patterns of G-CSF, to identify the impact of G-CSF prophylaxis on occurrence of neutropenia, and analyze potential risk factors of neutropenia in breast cancer and non-Hodgkin's lymphoma (NHL) patients. Methods: Eligible patients were those who diagnosed with NHL or breast cancer only, and initiated a new chemotherapy regimen in 2010 at a teaching hospital in Taiwan. All patients were followed until the planned chemotherapy courses were completed or one year after the first date of chemotherapy regimen. The duration and purpose of G-CSF use and the incidences of neutropenia and FN were evaluated by cycles. Logistic regression models and generalized estimating equations (GEE) were used to examine the association between CIN and other risk factors. Results: The incidences of neutropenia were 47% in 353 breast cancer patients (13% in 2776 cycles) and 79% in 72 NHL patients (40% in 433 cycles). The duration of G-CSF prophylaxis was 4.9±2.2 (mean±SD) days for primary prophylaxis and 3.7±1.5 days for secondary prophylaxis in breast cancer. The duration of G-CSF prophylaxis in NHL patients is shorter than that in breast cancer patients (primary prophylaxis: 2.9±0.3 days; secondary prophylaxis: 3.1±0.5 days). Multivariate logistic regression models found that G-CSF primary prophylaxis was associated with lower risk of neutropenia in breast cancer patient cycles (OR=0.49, 95%CI=0.30-0.80). No such effect was found when G-CSF was used for secondary prophylaxis (OR=0.96, 95%CI=0.67-1.37). Other risk factors of neutropenia included number of episodes of neutropenia during observational period and chemotherapy regimen with high risk of FN. In contrast, increased age was associated with lower risk of neutropenia. Results of GEE analysis suggested that chemotherapy regimen with high risk of FN was the only one risk factor of neutropenia in breast cancer patient cycles. However, we didn’t find the beneficial effects of primary or secondary prophylactic use of G-CSF in NHL patient cycles (OR=5.54, 95%CI=2.79-10.99; OR=2.30, 95%CI=1.23-4.30). Other risk factors of neutropenia included number of episodes of neutropenia during observational period, chemotherapy with high risk of FN, and lower hemoglobin (Hb). Similar results were found in GEE analysis. Compare to diffuse large B-cell lymphoma (DLBCL), marginal zone lymphoma (MALT) was associated with decreased risk of neutropenia. Conclusions: At this teaching hospital in Taiwan, the duration of G-CSF administration is shorter than the recommended. Several factors identified in this study can serve as good references for physicians to identify patients with high risk of neutropenia.