| 摘要: | 本篇研究包含兩個部分，在第一部分中我們優化一個精準的T細胞免疫組庫定序平台，在第二部分中我們將此平台應用於臨床研究。T細胞受體(T cell receptor, TCR)是T細胞表面的特異性受體，主要的功能是識別抗原並使T細胞活化，而T細胞免疫組庫(TCR repertoire)是對個體免疫系統中所有種類及數量的T細胞的概述，透過T細胞免疫組庫定序可以對健康人及病人做免疫反應的監控。在本研究中，我們發展了一套實驗及分析流程針對T細胞免疫組庫定序，它包含了唯一分子標記(unique molecular identifier, UMI)可以校正PCR反應產生的誤差，使我們更精確的測量T細胞克隆型的頻率。在參入分析(spiked-in analysis)中，我們將周邊血單核球細胞(PBMC)的RNA樣品添加不同比例的Jurkat T細胞株的RNA樣品，線性回歸分析的結果顯示添加的Jurkat T細胞株的RNA比例與定序後偵測到的Jurkat T細胞株克隆型的頻率有顯著的相關性(R2 = 0.99)。變異係數(coefficient of variation, C.V.)分析的結果顯示，在6個重複實驗的樣品中log2轉換後的克隆型頻率的變異係數為2.75%，這些結果顯示我們的T細胞免疫組庫定序的實驗及分析流程是高精準度、敏感度與重複性。
在過去十年中，大腸直腸癌(CRC)已為台灣惡性腫瘤發生率前三名的癌症。因此，我們運用T細胞免疫組庫定序平台想要了解在12個大腸直腸癌第三期的病人中，周邊血的T細胞組庫圖譜於手術及化療期間的變化情形，並觀察腫瘤及周邊血的T細胞組庫之間的關係。我們發現腫瘤T細胞組庫的TCR多樣性(diversity)比周邊血的T細胞組庫低，並且在治療的過程中周邊血的T細胞組庫的TCR多樣性有上升的趨勢。此外，大部分在周邊血中發現的腫瘤T細胞克隆型在腫瘤及周邊血的T細胞組庫內的數量皆比較高，而且會在治療的過程中長期存在於人體內，也發現在三個時間點都被觀察到的克隆型的比例在治療的過程中有下降的趨勢。此外，帶有相似TCR序列的T細胞克隆型在T細胞組庫內的比例比較高，並且腫瘤比起周邊血的T細胞組庫更容易發現帶有相似TCR序列的T細胞克隆型，表示腫瘤內具有較多T細胞傾向辨識相同或相似的抗原。總結來說，我們的研究說明了大腸直腸癌的病人在治療過程中T細胞組庫的變化以及抗腫瘤的免疫反應，這些發現突顯T細胞組庫的免疫監控可以作為臨床預後的指標。
This study consists of two sections. In section one, we optimized an accurate TCR-seq platform. In section two, we applied the TCR-seq platform to a clinical study. The T cell receptor (TCR) is a specific receptor on T cell surface that identifies antigen and activates T cells. TCR repertoire is an overview of the types and numbers of TCR that constitutes the individual adaptive immune system. TCR repertoire sequencing has been widely utilized in monitoring the adaptive immune responses of health and disease. In our study, we developed the experimental and computational pipeline for TCR sequencing. It contained unique molecular identifier (UMI) base correction to accurately measure the TCR frequency. In spiked-in analysis, the PBMC RNA samples spiked at different concentrations with RNA of Jurkat T cell. The linear regression analysis showed a significant correlation (R2 = 0.99) between the proportion of spiked-in Jurkat RNA and the frequency of detected Jurkat clones. In coefficient of variation (C.V.) analysis, the C.V. of log2 transformed frequency was 2.75% within the six replicated samples. Our TCR experimental method has accuracy, sensitivity and replicability.
Colorectal cancer (CRC) is the top three incidences of malignant tumors in the past ten years in Taiwan. Therefore, we applied the TCR sequencing platform to investigate the dynamic of peripheral TCR repertoires profiling during surgery and chemotherapy, and examined the relevance of tumor and peripheral blood repertoires in 12 patients with stage III CRC. We found that the diversity of TCR repertoires is lower in tumor than in peripheral blood. In peripheral blood, the TCR repertoire diversity increased during treatment. Furthermore, most of the tumor-associated clones (tumor clones found in peripheral blood) were abundant in both tumor and blood repertoires. During treatment, most of the tumor-associated clones remained exist in peripheral blood. The frequency of shared clones that were found at three timepoints of peripheral blood decreased during treatment. In addition, the clustered clones, formed by grouping TCRs with similar sequences, were highly abundant. The repertoires were highly clustered in tumor than blood, suggesting a high similarity in antigen specificity among tumor T cells. In summary, this study demonstrated the relevance between TCR repertoire dynamic and antitumor immune responses in CRC. These findings highlight that monitoring TCR repertoire may be useful as an indicator for clinical prognosis. |
