Back to overview

Great Cancer Therapy Relies on Accurate Tumor Samples

Personalized cancer therapy requires tumor samples

Cancerous tumors are unique to individual patients and constantly changing, making them very difficult to treat.

Although generic treatment options such as broad range chemotherapies and local radiation therapy are often effective to a limited extent, these are highly destructive for healthy tissues and therefore result in many and severe side-effects. Fortunately, over the last 2 decades, precision medicine has become available, enabling clinicians to choose therapies that can specifically target the cancer's unique molecular signatures. These treatments are highly specific and can therefore be more effective in destroying tumor cells, while leaving healthy tissues intact. Such treatments utilizing targeted therapies are often referred to as personalized medicine, and are becoming more common practice in the field of oncology.

In order to apply these targeted therapies effectively, clinicians need information on the tumor: what do the tumor cells look like on the outside and how may they respond to, for instance, hormone treatment? Such information can be obtained by taking a biopsy of a tumor: part of the tumor will be removed and can be analyzed in the lab using companion diagnostic tests to guide the best treatment.

73% of cancer drugs in development are personalized medicine (1)

Today unfortunately, standard biopsies are unreliable, often dangerous and may not truly represent the primary or metastatic tumor. If the information clinicians are receiving from current biopsy samples is inaccurate, how can they confidently choose the right treatment for their patients, at the right time? We have highly effective drugs and great diagnostics, but for them to work we also need equally adequate tumor samples for analysis.  

Every year, in the US alone, 1.5 billion dollars is spent due to complications arising following non-small cell lung cancer tissue biopsies

There are multiple different types of tissue biopsies routinely performed to guide treatment selection and monitoring of cancer. A needle biopsy, conducted during an imaging procedure such as an X-ray, ultrasound, CT or MRI scan, punctures and removes tumor tissue beneath the skin (2). In more difficult to reach cancers, such as lung cancer, more invasive and often dangerous biopsies are sometimes required. Transbronchial biopsies involve the use of a bronchoscope, a tube with a telescope on the end, that enters the lungs through the main airways (3). Even more invasive, thoracoscopic biopsies are performed under general anaesthesia, in which an endoscope enters the chest cavity through the chest wall. Surgical tools can enter through the bronchoscope or endoscope and are used to remove small pieces of lung tumor tissue. This tissue is then analyzed, usually under a microscope, to detect certain druggable targets to help guide treatment. Tissue biopsies are currently the gold standard in clinical practice. Not because of their safety or effectiveness, but simply because there are no reliable alternatives. In fact, taking a tissue biopsy can involve high risks. Shockingly, in the USA alone, 1.5 billion USD is spent on complications associated with tissue biopsies for lung cancer every year (4-6)*. To prevent these complications, as well as the associated suffering and costs, we need a less invasive approach for tumor sampling.

Cancer diagnostics are only as good as the cancer sample

Lung cancer causes more deaths per year than any other cancer, accounting for 1.8 million deaths globally in 2020 (7). Non-small cell lung cancer (NSCLC) makes up 80% of lung cancer cases. There are a multitude of different treatment options for NSCLC and it is crucial that the correct and most effective treatment is chosen to ensure patients receive the best possible outcome. While tissue biopsies remain the primary type of biopsy to guide lung cancer treatment, surprisingly only 18% of NSCLC patient’s tissue biopsy will provide adequate samples for complete tissue genotyping for all of the eight-guideline FDA recommended genomic biomarkers (8).

When choosing treatment, there are several FDA cleared or approved companion diagnostic kits / tests available, allowing clinicians to analyze a patient's tumor for specific markers or drug resistant mutations following a biopsy (9). These tests predominantly detect mutations, such as in the epidermal growth factor receptor (EGFR), or the presence of particular proteins expressed on tumor cells, including anaplastic lymphoma kinase (ALK) and programmed death ligand-1 (PD-L1). Recent advances in next generation sequencing (NGS) technologies, otherwise known as massive parallel sequencing, have resulted in the development of commercially available tests, such as the Oncomine Dx target test, that can detect dozens of oncogenic gene variants in one tissue sample.

These tests are very reliable, however the tumour samples acquired via tissue biopsy very often do not capture the full heterogeneity of the tumor, but rather a snapshot of a small section. The test might be right, but the sample might not be, resulting in potentially lifesaving treatments being withheld from patients, or being overprescribed.

A single cancer sample is not enough

In addition to being heterogenous, tumors are constantly changing. However, as tissue biopsies are invasive and potentially dangerous, this technique cannot be used for frequent tumor sampling. Tumor plasticity is clearly illustrated by PD-L1, a protein expressed on the surface of certain tumor cells, which can indicate if a NSCLC patient may respond to immunotherapy or not. Currently, 41% of patient's tumor biopsies invalidate over the course of their treatment, due to changes in PD-L1 expression, either by increasing or decreasing (10-14). Not identifying tumor PD-L1 expression changes can result in potentially 41% of patients being given the wrong treatment, for a cancer type that causes the most human mortality. The invasiveness and turnaround time required for tissue biopsies means that frequent tumor biopsies are simply infeasible. Advancements in personalized medicine are therefore redundant if cancer patients are being prescribed the wrong treatment based on outdated samples.

Problems at every step of the tissue biopsy process

The process of determining a patient's PD-L1 status, and other tumor molecular signatures, via tissue biopsy, traditionally involves a multi-step process. The patient's tumor is sampled by a clinician and then processed in a lab. The gold standard processing method involves immunohistochemical staining to visualise PD-L1 expression. The processed sample is then analyzed under a microscope through which an operator will determine the percentage of PD-L1 expressed on the patient's tumor. This percentage will then guide patient treatment options, stratifying them based on the treatment they will or won't receive. It is therefore crucial patients are placed into the right treatment group.

An analysis of literature and clinical trial data comparing matching tumor samples revealed discrepancies at every step of the process. When sampling the same tumor, depending on the method used to extract tissue, it was found that 15% of methods (surgical resection vs other) result in differing PD-L1 expression results, meaning 15% of patients would be placed in the wrong treatment group (20, 21). Depending on the particular tumor sampled in the same patient, whether that be a primary tumor, paired primary or distant metastases, 31% will differ in PD-L1 expression (22, 23). There are four FDA approved immunohistochemistry assays used to analyze a patient's tumor sample. It was found that when using the different assays on the same patient tumor sample, the PD-L1 expression results will differ 18% of the time (24, 25). Furthermore, during the analysis, when comparing the results of different operators analysing the same tissue, 12% of operators will disagree on the percentage PD-L1 expression (26-29).

An overview of the percentage of NSCLC patients being misclassified into the wrong immunotherapy treatment group at every step of the PD-L1 tumor biopsy process.

Liquid biopsies are part of the answer

The levels of ctDNA in a standard blood sample are so low that often machine learning algorithms are required to detect a signal following NGS analysis.

Liquid biopsies are a type of biopsy based on blood samples that are analysed in vitro for circulating bioanalytes, such as circulating tumor cells (CTCs), circulating tumor DNA (ctDNA) or extracellular vesicles (EVs) / exosomes. They are becoming increasingly popular as potential candidates to replace current tissue biopsies. Several liquid biopsy based tests are commercially available as companion diagnostic tests to guide the treatment of multiple different cancers, including NSCLC. Currently the most popular liquid biopsy test is profiling of ctDNA from blood samples. CtDNA is a type of cell-free DNA (cfDNA) that has been shed from a patient’s tumor cells, and circulates in the blood stream (15). PCR-based detection and profiling of specific genomic signatures, such as methylation patterns on ctDNA, has led to the development of commercial liquid biopsy companion diagnostic tests for multiple cancers (16).

Next Generation Sequencing (NGS) technology is now being utilized in commercially available liquid biopsy companion diagnostic tests for NSCLC. FoundationOne’s Liquid CDx test is the first companion diagnostic liquid biopsy test to be granted FDA approval for multiple cancers including NSCLC, breast, ovarian and prostate. ArcherDx, recently acquired by Invitae, were granted FDA Breakthrough Status for Stratafide, a pan-cancer NGS diagnostic platform, that combines both solid tumor tissue and liquid biopsy companion diagnostic testing into one assay. The test was expected to be approved in March 2022, but FDA feedback has resulted in delays. It is not known when, or even if, the world’s first pan-cancer NGS tissue and liquid biopsy companion diagnostic test will reach the market.

While advances in liquid biopsy ctDNA NGS analysis have been major steps forward in the oncology companion diagnostic field, the sensitivity of these tests remain low, simply because so little ctDNA exists in a blood sample. This leads to a significant number of false negative results (17) prompting the FDA, as can be seen from the diagnostic algorithm below, to always require a tissue biopsy following a negative liquid biopsy result, making these types of test partially redundant. More sensitive liquid biopsy NGS technologies are currently being developed for lung cancer to solve this problem, but we can’t continue to rely on tissue biopsies.

Proposed diagnostic algorithm for NSCLC from the International Association for the Study of Lung Cancer. If a tissue sample is available they recommend a "sequential approach" in which cfDNA analysis is used if a tissue sample doesn't provide adequate tissue for genotyping, or a complementary approach where tissue and liquid samples are analysed concurrently. If no tissue sample is available a "plasma first approach" is recommended, followed by a tissue biopsy if the liquid biopsy was a negative result (18).

Can flow biopsies solve these problems?

As tissue biopsies have multiple major disadvantages and liquid biopsies are currently not sensitive enough, perhaps another type of biopsy that enriches circulating tumor cells (CTCs) can provide a solution. CTCs are shed from a patient's tumor and circulate throughout their bloodstream. These tumor derived cells can be analyzed using the same techniques as ctDNA to give the same genomic, epigenomic and transcriptomic information about a patient's tumor. Furthermore, unlike ctDNA, proteomic analysis of CTCs can give information about specific proteins expressed on tumor cells, such as PD-L1. So if the downstream analytical capabilities with CTCs are greater than that of ctDNA, why are there a lack of liquid biopsy tests utilising them? The only major issue with these types of cells is their low concentration in blood, particularly in non-metastatic patients (19), which makes a proper analysis impossible follwoing a single blood draw.

Idris Oncology hopes to solve the problems associated with tissue and liquid biopsies with their flow biopsy technology. Using a wire with a specialized coating, CTCs can be efficiently captured directly from a patient's blood stream in a non-invasive manner. This allows for the sampling of liters of blood, rather than the 7 to 10 ml in a regular blood sample. This will allow for efficient enrichment of CTCs that can be used in multiple diagnostic applications, without the need for invasive and unreliable tissue biopsies. In the not-so-distant future, we may see a world in which flow and liquid biopsy technologies work in parallel to accurately analyze tumor samples safely, efficiently and accurately, ensuring patients always get the right treatment, at the right time.


(1) Krzyszczyk, P., Acevedo, A., Davidoff, E., et al, 2018. The growing role of precision and personalized medicine for cancer treatment. TECHNOLOGY, 06(03n04), pp.79-100.

(2) Pritzker, K. and Nieminen, H., 2019. Needle Biopsy Adequacy in the Era of Precision Medicine and Value-Based Health Care. Archives of Pathology & Laboratory Medicine, 143(11), pp.1399-1415.

(3) Udagawa, H., Kirita, K., Naito, T., et al, 2020. Feasibility and utility of transbronchial cryobiopsy in precision medicine for lung cancer: Prospective single‐arm study. Cancer Science, 111(7), pp.2488-2498.

(4) Kelly, R., Turner, R., Chen, Y., et al. 2019. Complications and Economic Burden AssociatedWith Obtaining Tissue forDiagnosis and Molecular Analysis in Patients With Non–Small-Cell Lung Cancer in the United States.Journal of OncologyPractice, 15(8), pp.e717-e727.

(5) Zhang, Y., Shi, L., Simoff, M., JWagner, O., et al. 2020. Biopsy frequency and complications among lung cancerpatientsin the United States.Lung Cancer Management, 9(4), p.LMT40.

(6) Chiu, Y., Kao, Y., Simoff, M., et al .2021. Costs of Biopsy andin Patients with Lung Cancer.ClinicoEconomicsand Outcomes Research, Volume 13, pp.191-200.

(7) 2022. Cancer. [online] Available at: <> [Accessed January 2022].

(8) Leighl, N., Page, R., et al. 2019. Clinical Utility of Comprehensive Cell-free DNA Analysis to Identify Genomic Biomarkers in Patients with Newly Diagnosed Metastatic Non–small Cell Lung Cancer. Clinical Cancer Research, 25(15), pp.4691-4700.

(9) U.S. Food and Drug Administration. 2022. List of Cleared or Approved Companion Diagnostic Devices. [online] Available at: <> [Accessed January 2022].

(10) Hong, L., Negrao, M., et al. 2020. Programmed Death-Ligand 1 Heterogeneity and Its Impact on Benefit From Immune Checkpoint Inhibitors in NSCLC. Journal of Thoracic Oncology, 15(9), pp.1449-1459.

(11) Omori, S., Kenmotsu, H.,et al. 2018. Changes in programmed death ligand 1 expression in non-small cell lung cancer patients who received anticancer treatments. International Journal of Clinical Oncology, 23(6), pp.1052-1059.

(12) Cho, J., Sorensen, S.,et al. 2017. Programmed Death Ligand 1 Expression in Paired Non–Small Cell Lung Cancer Tumor Samples. Clinical Lung Cancer, 18(6), pp.e473-e479.

(13) Naso, J., Banyi, N., et al. 2021. Discordance in PD-L1 scores on repeat testing of non-small cell lung carcinomas. Cancer Treatment and Research Communications, 27, p.100353.

(14) Shin, J., Chung, J., et al. 2019. Effect of Platinum-Based Chemotherapy on PD-L1 Expression on Tumor Cells in Non-small Cell Lung Cancer. Cancer Research and Treatment, 51(3), pp.1086-1097

(15) Kwapisz, D., 2017. The first liquid biopsy test approved. Is it a new era of mutation testing for non-small cell lung cancer?. Annals of Translational Medicine, 5(3), pp.46-46.

(16) Neumann, M., Bender, S., Krahn, T., et al., 2018. ctDNA and CTCs in Liquid Biopsy – Current Status and Where We Need to Progress. Computational and Structural Biotechnology Journal, 16, pp.190-195.

(17) Paweletz, C., Lau, C. and Oxnard, G., 2019. Does Testing Error Underlie Liquid Biopsy Discordance?. JCO Precision Oncology, (3), pp.1-3.

(18) Rolfo, C., Mack, P., et al. 2021. Liquid Biopsy for Advanced NSCLC: A Consensus Statement From the International Association for the Study of Lung Cancer. Journal of Thoracic Oncology, 16(10), pp.1647-1662.

(19) Rushton, A., Nteliopoulos, G., et al. 2021. A Review of Circulating Tumour Cell Enrichment Technologies. Cancers, 13(5), p.970.

(20) Kitazono, S., Fujiwara, Y., et al. 2015. Reliability of Small Biopsy Samples Compared With Resected Specimens for the Determination of Programmed Death-Ligand 1 Expression in Non–Small-Cell Lung Cancer. Clinical Lung Cancer, 16(5), pp.385-390.

(21) Li, C., Huang, C., et al. 2017. Comparison of 22C3 PD-L1 Expression between Surgically Resected Specimens and Paired Tissue Microarrays in Non–Small Cell Lung Cancer. Journal of Thoracic Oncology, 12(10), pp.1536-1543.

(22) Kim, S., Koh, J., et al. 2017. Comparative analysis of PD-L1 expression between primary and metastatic pulmonary adenocarcinomas. European Journal of Cancer, 75, pp.141-149.

(23) Luo, L., Luo, X., et al. 2020. Consistency Analysis of Programmed Death-Ligand 1 Expression between Primary and Metastatic Non-Small Cell Lung Cancer: A Retrospective Study. Journal of Cancer, 11(4), pp.974-982.

(24) FUJIMOTO, D., YAMASHITA, D., et al. 2018. Comparison of PD-L1 Assays in Nonsmall Cell Lung Cancer: 22C3 pharmDx and SP263. Anticancer Research, 38(12), pp.6891-6895.

(25) Kim, H., Kwon, H., et al. 2017. PD-L1 immunohistochemical assays for assessment of therapeutic strategies involving immune checkpoint inhibitors in nonsmall cell lung cancer: a comparative study. Oncotarget, 8(58), pp.98524-98532.

(26) Cooper, W., Russell, P., et al. 2017. Intra- and Interobserver Reproducibility Assessment of PD-L1 Biomarker in Non–Small Cell Lung Cancer. Clinical Cancer Research, 23(16), pp.4569-4577.

(27) Brunnström, H., Johansson, A., et al. 2017. PD-L1 immunohistochemistry in clinical diagnostics of lung cancer: inter-pathologist variability is higher than assay variability. Modern Pathology, 30(10), pp.1411-1421.

(28) Rehman, J., Han, G., et al. 2016. Quantitative and pathologist-read comparison of the heterogeneity of programmed death-ligand 1 (PD-L1) expression in nonsmall cell lung cancer. Modern Pathology, 30(3), pp.340-349.

(29) Rimm, D., Han, G., et al. 2017. A Prospective, Multi-institutional, PathologistBased Assessment of 4 Immunohistochemistry Assays for PD-L1 Expression in Non–Small Cell Lung Cancer. JAMA Oncology, 3(8), p.1051.

* The complication costs are based on the mean yearly cost of biopsies minus the median yearly cost of biopsies according to source (4-6) and the amount of NSCLC patients according to the National Cancer Institute.

September 7, 2022

Conor Simpson
Business Development | Market Analyst
Conor Simpson

More blogs