Preanalytical aspects: A neglected issue

Abstract

Considerable research data is available that demonstrate that tissues are under tremendous biological stress when surgically separated from the body. This stress significantly changes gene and protein expression profiles, including activation or inhibition of signalling pathways and their receptors. Many of those are possibly involved in growth regulation and might serve as targets or stratification markers for new drugs. Factors that affect tissue dependent cancer research for target discovery and drug development include drug treatment and anesthesia of patients before surgical tissue removal, intrasurgical ischemia by ligation of main arteries, “cold” ischemia, i.e. the time interval between surgical removal and fixation of tissue, location of tumor biopsy within a given tumor, processing of tissue and fixation protocols and the availability of comprehensive clinical data. Controlled and rapid tissue processing is a prerequisite for understanding biological differences of patient tumors and to utilize these findings (e.g., cancer pathway activity) for targeted molecular therapies.

Key Words::

• Biobank
• tissue quality
• cancer
• individualized therapy

Introduction

Cancer is a highly heterogeneous disease which has become the main cause of death in the Western World. Since Richard Nixon declared the war against cancer in 1972, our knowledge of the molecular basis of cancer has grown tremendously. However, basic findings made in cancer cell lines have been barely translated, so far, into improved therapy and outcome. This achievement remains to be the main task. Therefore, hope for curing cancer relies on understanding molecular differences in patients’ tumors and on developing molecular therapeutics together with biomarkers for stratifying patients to apply targeted therapeutics accordingly.

Indivumed was founded in 2002 to support this task and to help in developing individualized cancer medicine. As an essential basis for achieving this goal it was understood that a tumor biobank and clinical infrastructure is needed that enables the understanding of differences in cancer patients and the translation of our molecular knowledge into new, targeted therapies. The availability of comprehensive sets of biospecimens (tumor and normal tissue, plasma, serum, blood cells, urine) from each patient together with complete clinical data and data about biospecimen processing has been a major goal, together with the assurance that biospecimens, in particular, tissues, reflect the molecular pattern of cells as close to the reality in the living body as possible.

This fundamental understanding of translational research is still a widely neglected issue, even though in 2004 the US National Cancer Institute (NCI) founded the “Office Biorepositories and Biospecimen Research” (OBBR) because the lack of high quality biobanks is regarded as the number one road block on the way to cure cancer. Furthermore, the OBBR declared publicly that in the past 15 years the NCI has spent several billion USD in funding of projects which turned out to be useless because tissue quality was neglected.

Considerable research data, published and unpublished, is available that demonstrate that tissues are under tremendous biological stress when surgically separated from the body. This stress significantly changes gene and protein expression profiles, including activation or inhibition of signalling pathways and their receptors. Many of those are possibly involved in growth regulation and might serve as targets or stratification markers for new drugs. Indivumed has put (and continues to do so) a significant amount of effort into characterizing the impact of these preanalytical factors on tissue research data. In part, this research is performed in collaboration with the OBBR at the NCI. In the following, important factors impacting molecular data are summarized.

Drug treatment and anaesthesia of patients before surgical tissue removal

According to Indivumed's data base, which accurately records all drugs and their doses intrasurgically given to patients, approximately 112 different compounds in various combinations are routinely applied. Naturally, this includes numerous drugs which interact with metabolic pathways and cell function. The knowledge about application of these compounds becomes extremely valuable in discovery projects of biomarkers and pathways. One example is a discovery project of colon cancer biomarkers by Indivumed. By correlating potential biomarker candidates with intrasurgical drug treatment, four proteins could be identified which clearly indicate application of the drug Atropin as the reason for differential expression, but not for cancer.

Intrasurgical ischemia by ligation of main arteries

Ligation of main arteries during surgery results in hypoxia of cancer and normal tissue and subsequently stress-induced cellular dysregulation. A pilot project of Indivumed and the OBBR analysed tissue from colon cancer patients resected by left hemicolectomy [1]. This type of surgery includes a ligation of an artery that provides ∼90% of the blood flow to the colon. Indivumed's data base allowed identifying subgroups of patients with different time intervals between artery ligation and completion of resection, while postsurgery ischemia time was the same in all cases (10 minutes). cDNA microarray gene expression analysis of laser capture microdissected tumor/epithelial cells identified numerous genes showing expression levels that correlate with intrasurgical ischemia time. A prospective trial has recently been initiated to confirm these data and to identify those molecules that reflect unsecure data points and, possibly, to identify markers that allow to determine tissue quality and to correct for ischemia related changes.

“Cold” ischemia, i.e. the time interval between surgical removal and fixation of tissue

The first two factors can only be documented and then used, for example, to build accurate comparative groups or to evaluate research data at a later time. In contrast, the cold ischemia time, i.e., the time between completion of surgery and fixation of tissues, as well as factors 4–7 listed below, can be controlled by accurate tissue processing that starts ideally in the surgery unit. Several studies, including studies performed by Indivumed, clearly indicate that the time between tumor resection and tissue fixation is essential to obtain reliable research data [2,3]. In particular, data quality of phosphoprotein concentration measurements and determination of pathway regulating molecules highly depends on the speed and accuracy of tissue processing following surgical resection. We found that in colon tissue as soon as 5 minutes after surgical resection gene and protein expression data start to change. After 10–15 minutes > 10% of all proteins and genes are differentially expressed, amounting to 25% of all genes and proteins within 30 minutes [3]. More recently, we determined phosphoprotein levels of various key molecules which are also included as stratification biomarker in clinical trials. It becomes obvious that phosphorylation of proteins such as mTOR (= mammalian Target of Rapamycin) and MAPK (= Mitogen-Activated Protein Kinases) are significantly affected by cold ischemia. Within 10–15 minutes phosphorylation significantly changed in a majority of cases, resulting in increasing pmTOR concentrations with high expression levels after 45 minutes. MAPK showed a more heterogeneous picture with cases of increasing phosphorylation concentrations and other cases with decreasing phosphorylation concentrations. We conclude from these data that without knowledge of the exact cold ischemia time, protein and RNA research, especially in target discovery projects, is extremely risky. Studies about pathway activity, phosphoprotein expression and interaction of signalling molecules are worthless if cold tissue ischemia is not controlled below 10–15 minutes.

Location of tumour biopsy within a given tumor

Tumors usually contain two biologically different areas: an invasive and usually well nutritioned periphery and a hypoxic, necrotic center. This is reflected by protein and gene expression studies that clearly indicate a high variability between these two tumour areas. When tumor and normal tissues are compared, 40% of cancer related proteins (as determined by mass-spectrometry) were found either in the center or in the periphery. Only 60% of differentially expressed proteins were found in both tumour areas. Hypoxic conditions seem to play a role as suggested by the fact that VEGF expression is significantly higher in the tumor center than in the tumor periphery. In conclusion, it is helpful to know about the origin of tumor tissues and to collect centre and periphery areas separately.

Comparative tissues to define cancer related molecules

Discovery projects to identify cancer related molecules for diagnosis and therapy are usually based on comparative studies of tumor and corresponding normal tissue. Naturally, normal organ tissue is composed of different cell types and possibly tissue of different origins (e.g., epithelial cells and muscle within a colon wall). In contrast, cancer tissue is composed of cancer cells (which are derived from epithelial cells) and some stroma cells. However, muscle cells, for example, are not present in a colon cancer. This needs to be considered when comparative tissue studies are performed. In colon cancer research, tumor tissue should be solely compared with the mucosa, not the entire colon wall, a consideration which has been neglected in numerous studies. Consequently, tissue sampling should consider this fact by collecting the corresponding tissue and not necessarily the corresponding normal organ.

Processing of tissue and fixation

Formalin fixation is the most common approach for tissue fixation. Standard pathology samples are usually processed by fixation of the entire resected organ in a bowl filled with formalin. Subsequently, the fixation process is highly variable because formalin diffusion into the tissue takes time. Normal and tumor tissues are only fixed within a variable and wide time interval. Both cold ischemia and the ratio of tissue vs. formalin volumes have potentially a high impact on immunostaining results with respect to the image quality and, more importantly, the interpretation of protein expression data. Therefore, we prefer to process tissue instantly after resection in same-sized tissue blocks to assure an identical and rapid fixation process of tumor and normal tissue.

Fluid preparation

Blood and urine are also affected by their separation from the body. In contrast to tissue acquisition, however, standardized processing of fluids is much better established. Instant cooling to preserve the molecular status by reducing the metabolic activity of blood cells can (and should) be easily performed. Overall, processing of cooled blood in cells and plasma within 3–4 hours has been found in most studies to preserve DNA and protein integrity. Longer processing times can result in DNA, RNA and protein degradation.

Primary cells and organoid tissue cultures

Besides archived biospecimens, primary cancer cells and viable tissue cultures represent model systems that help translation of biological findings in clinically meaningful results. We have studied molecular variability among instantly fixed tumor tissue, primary cancer cells isolated from tumors and organoid cultures, i.e. 200–400 µm slices of tissues. Within 24–48 h, we saw rapidly decreasing molecular comparability between primary cells and the original tumor. In contrast, intact tissue cultures conserved the molecular pattern for at least 72 h of cultivation time. This provides a sufficient time window to study drug effects in a clinically valuable model that helps understanding the heterogeneity among patients’ tumors.

Summary

The availability and consequent use of a comprehensive data base and biobank with high quality tissues (i.e., freezing/fixation of similar sized tissue blocks in less than 10 minutes) has a significant impact on the effectiveness of target discovery and the development of new drugs and their companion diagnostics. It needs to be understood in the research community that an initial investment in quality is essential to an improvement in cancer research and drug development.

Having been recognized by the NCI as a sole source because of its unique quality standard and clinical infrastructure, Indivumed can serve as an example of how science-guided biobanking should be performed. Indivumed has implemented a clinical infrastructure and an approach for biospecimen collection which records all known factors that might impact research data and controls and standardizes biospecimen collection processes as much as possible. This demands a level of control and processing which is identical to an industrial level of product control. Consequently, Indivumed is running its processes under ISO 9001:2008 norm.

As a result of such costly efforts, a platform exists for analysis of target expression and pathway activity in tissue reflecting the molecular reality in patients and for the determination of molecular drug effects in short-term cultivated tissues. The platform provides an excellent basis for accelerating the preclinical and clinical development process and, subsequently, reducing R&D cost.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.