关于不同分子亚型的大肠癌总生存期（OS）和无进展生存期（PFS）有何不同，Tejpar博士认为有几点需要考虑。大肠癌分子分型是为了诊断，不是为了预测预后的差异。不同的亚型有不同的生物学特征，因此预后也不同。如果要以分型确定预后，就要换一种完全不同的方法。

　　Tejpar博士认为主要是以下两种情况需要确定肿瘤的分子生物学差异。一是为了预测复发（RFS）风险。其二是结肠癌新辅助治疗。II期或III期的原发肿瘤切除后有转移的风险，在这种情况下CMS4能增加复发风险率，CMS4可用作新靶向药物的靶标。根据基因表达进行分子分型，不仅要检测信号通路还要检测替代指标。结直肠癌的生物学差异对复发后的生存期影响很大。结直肠癌分子分型对化疗敏感性、复发风险等有预测作用。故不仅要理解肿瘤生物学差异，还要找出这些差异产生机制，但这种工作即困难又不太可行，需要大量的临床前观察，而这是Tejpar博士目前的研究重点之一。

　　结直肠癌生物亚型可影响靶向治疗和化疗反应、临床护理、患者的管理，但是还需要大量临床研究验证。结直肠癌在生物学上差异巨大，其对常用靶向治疗的反应也不同。因此结直肠癌需要把患者进行分子分型，进行不同的药物试验，检测不同药物的疗效，并进行对照研究。

　　结肠癌一般是根据基因表达、microRNA、蛋白质进行分型。目前没有单一标志物可以预测这个复杂疾病的治疗，找出预测疗效的生物标志物需要大量的工作，同时还需要临床试验来验证。

访谈原文

　　Oncology Frontier: Please discuss how overall survival (OS) and progression-free survival (PFS) are differentiated among the various molecular subtypes of colorectal cancer?

　　《肿瘤瞭望》：不同分子亚型的大肠癌总生存期（OS）和无进展生存期（PFS）有何不同？

　　Dr Tejpar: There are several points to consider. Subtypes were generated in a diagnostic way， not any way that reflects prognostic differences. That is very important. Subtypes were generated to show differing biology in colon cancer and once those subtypes were identified based on that differing biology， we checked if there were also prognostic differences. If you wanted a prognostic classification， you would need to use a completely different approach. When we identified these basic biological differences in colon cancer， we look at how they behave in two settings mainly. One setting is the risk for relapse (RFS) once there is a resected primary tumor in stage II or III so that we know the risk for developing metastases. We do indeed see a subtype and in this case it is CMS4. That really makes a difference in terms of the risk of relapse. So now we have a molecularly diagnostically defined subtype which contains biology that is associated with the risk for relapse and this can now be utilized for new drug targets.

　　That adjuvant area is very important and something we need to work on in colon cancer. When we were making the subtypes in this case they were generated by gene expression which use a lot of variables and associating them by looking at patterns which means that you are not really capturing the underlying pathway but maybe even looking at surrogate markers. These tell you that this patient is different in terms of those features from another patient. What we are doing now and what we are looking into is trying to give some functional understanding to these signals which help us to define the patients. We have signals that define our patients as being biologically different and in this case also prognostically different， but now we have to go and find what exactly is mechanistically going on in a patient that is expressed by these features. That is difficult because you need to have the key to unlock a phenotypic expression where we need to know what does it mean when particular genes are turned on or off. It is difficult but not unfeasible. It takes a lot of preclinical work to gather these observations and that is what we will be focusing on over the next year. When we consider the other outcome， we could look at overall survival or also survival-after-relapse. That is quite a clean variable because time zero is when the patient becomes metastatic and then we look at which subgroup influences outcomes in that metastatic setting. Indeed， again we see the subgroups separate in terms of survival-after-relapse. Not unexpectedly， the biology that we see is important in survival-after-relapse is not the same biology as the one that confers the risk for relapse to begin with. That makes a lot of sense and gives us clues for if you are working on chemosensitivity and survival in the metastatic setting， you are going to be addressing other things other than those involved in the risk for relapse. So sometimes， subtypes are prognostic and give us hints on relevant clinical features (such as risk of relapse and unresponsiveness to chemotherapy) where we can zoom in to find the molecular mechanisms.

　　Tejpar博士：有几点需要考虑。大肠癌分子分型是为了诊断，不是为了预测预后的差异。不同的亚型有不同的生物学特征，因此预后也不同。如果要以分型确定预后，就要换一种完全不同的方法。主要是以下两种情况需要确定肿瘤的生物学差异。一是为了预测复发（RFS）风险，II期或III期的原发肿瘤切除后有转移的风险，在这种情况下CMS4会增加复发风险，其可用于新靶向药物的靶标。其二是结肠癌新辅助治疗。根据基因表达进行分子分型，不仅要检测信号通路还要检测替代指标。这种情况下不仅要理解肿瘤生物学差异，还要找出这些差异产生机制，但这种工作即困难又不太可行，需要大量的临床前观察，而这就是我们明年工作的重点。生物学差异对复发后的生存期影响很大。结直肠癌分子分型对化疗敏感性、复发风险等有预测作用，而我们正努力找出其中的机制。