#213 ‒ Liquid biopsies and cancer detection | Max Diehn, M.D. Ph.D.

Jul 11, 2022 Episode Page ↗
Overview

Max Diehn, Professor of Radiation Oncology at Stanford, discusses the history and future of liquid biopsies for cancer detection. He explains the importance of diagnostic test metrics like sensitivity and specificity, and how cell-free DNA analysis is revolutionizing early cancer detection and guiding adjuvant therapies.

At a Glance
12 Insights
2h 6m Duration
15 Topics
9 Concepts

Deep Dive Analysis

Max Diehn's Background and MD-PhD Journey

Decision to Specialize in Radiation Oncology

Motivation for Developing Liquid Biopsies

Limitations of Traditional Cancer Imaging and Detection

Historical Blood-Based Cancer Markers (Proteins)

Understanding Diagnostic Test Metrics: Sensitivity and Specificity

Current State and Challenges of Lung Cancer

Low-Dose CT Scans for Lung Cancer Screening

Shift from Circulating Tumor Cells to Cell-Free DNA

Detecting Cancer Mutations in Cell-Free DNA

Cell-Free RNA and Other Informative Signatures

Challenges and Promise of Pan-Cancer Screening with Liquid Biopsies

FDA-Approved Liquid Biopsies and Clinical Use Cases

Future of Liquid Biopsies: Guiding Adjuvant Therapy

The 'Holy Grail' of Cancer Screening and its Challenges

DNA Microarrays

A technology invented by Pat Brown that enabled the simultaneous measurement of the expression of virtually all genes in the genome. This allowed researchers to study tens of thousands of genes in a single experiment, opening new fields in biology.

Sensitivity

Also known as the true positive rate, it is the probability that a diagnostic test will correctly identify individuals who have a particular condition. For example, if 100 cancer patients are tested, sensitivity measures how many are correctly identified as positive.

Specificity

The inverse of sensitivity, it is the probability that a diagnostic test will correctly identify individuals who do not have a particular condition. If 100 healthy individuals are tested, specificity measures how many are correctly identified as negative.

Cell-Free DNA (cfDNA)

DNA molecules found circulating in the blood plasma, outside of cells. These short, double-stranded DNA fragments are largely protected by histones and are contributed by most tissues in the body, including healthy and cancerous cells.

Circulating Tumor DNA (ctDNA)

A specific sub-fraction of cell-free DNA that originates from cancer cells. It contains unique mutations present in the tumor and serves as a highly specific molecular marker for the presence of cancer.

Apoptosis

A mode of programmed cell death, a controlled 'suicide' mechanism within cells that is crucial for development and for eliminating damaged or unnecessary cells. This process involves the systematic chopping up of DNA and proteins.

Clonal Hematopoiesis

A phenomenon in older individuals where mutations accumulate in white blood cells over time. These mutations can be released into the plasma as cell-free DNA, necessitating their differentiation from tumor-derived mutations in liquid biopsy screening.

DNA Methylation

A chemical modification to DNA molecules, specifically the addition of a methyl group, which can influence gene expression. Different cell types and tissues exhibit distinct methylation patterns, making them potential markers for identifying the tissue of origin of cell-free DNA.

Minimal Residual Disease (MRD)

Microscopic cancer cells that persist in the body after initial treatment (like surgery or radiation) but are too small to be detected by conventional imaging. Liquid biopsies aim to detect ctDNA from MRD to guide further adjuvant therapies.

?
What is the primary motivation behind developing liquid biopsies for cancer?

The motivation stems from the frustration of not being able to diagnose cancer recurrence earlier in patients who have undergone treatment, as traditional imaging methods can only detect tumors once they reach about a billion cells.

?
Why are traditional protein biomarkers like PSA, CEA, and CA-19-9 limited for cancer screening?

These protein biomarkers lack specificity, meaning normal cells can also produce them, making it difficult to distinguish between low levels produced by normal cells versus those indicative of cancer.

?
What are the limitations of using circulating tumor cells (CTCs) for cancer detection?

CTCs are not very abundant, difficult to purify, require rapid sample processing, and some methods lack specificity, picking up non-cancerous epithelial cells in healthy individuals.

?
How do liquid biopsies using cell-free DNA (cfDNA) identify cancer?

They focus on detecting mutations unique to cancer cells within the short fragments of cfDNA circulating in the blood, as these mutations are present only in cancer cells and not in the patient's normal cells.

?
What is the distinction between cell-free DNA (cfDNA) and circulating tumor DNA (ctDNA)?

Cell-free DNA refers to the total DNA in circulation, including both healthy and cancer-derived DNA, while circulating tumor DNA is specifically the sub-fraction of cfDNA that originates from tumor cells.

?
Why is it challenging to use liquid biopsies for early cancer screening in the general population?

Even with high sensitivity and specificity, if the pre-test probability of cancer in a low-risk individual is very low (e.g., 1%), a positive result is still more likely to be a false positive, leading to unnecessary follow-up and anxiety.

?
What is the current primary clinical use for FDA-approved liquid biopsies like Guardant?

These tests are primarily used in patients with metastatic cancer to identify specific actionable mutations in their tumors, which can guide targeted drug therapies without needing an invasive tissue biopsy.

?
How does the FDA regulate liquid biopsy tests, and what is the 'CLIA' pathway?

The FDA provides approval for some tests, but many diagnostic companies use the CLIA (Clinical Laboratory Improvement Amendments) pathway, where the lab itself is regulated for appropriate procedures, allowing them to offer tests to patients without individual FDA review.

?
How can liquid biopsies guide adjuvant therapy in cancer patients?

By detecting minimal residual disease (MRD) through ctDNA after initial treatment, liquid biopsies can identify patients with microscopic cancer cells who are at high risk of recurrence, allowing for targeted adjuvant therapies to potentially cure more patients at their lowest tumor burden.

1. Understand Diagnostic Test Interpretation

Learn sensitivity, specificity, prevalence, positive predictive value (PPV), and negative predictive value (NPV) to accurately interpret diagnostic test results, especially for cancer screening. Utilize available calculators or apps to plug in these values and understand how they influence a test’s utility.

2. Prioritize Early Cancer Detection

Recognize that earlier cancer detection, which leads to lower tumor burden and reduced genetic heterogeneity, consistently results in better patient outcomes and increased chances of successful treatment. This fundamental principle underscores the critical importance of catching cancer as early as possible.

3. Demand Rigorous Screening Test Trials

Advocate for and support large, randomized clinical trials that prove new cancer screening tests, such as liquid biopsies, decrease cancer-specific mortality before widespread adoption. This ensures that new technologies genuinely save lives and do not merely add healthcare costs or patient anxiety.

4. Guide Adjuvant Therapy with Liquid Biopsies

If you are an early-stage cancer patient who has completed initial treatment (surgery/radiation), discuss with your doctor the option of using highly sensitive liquid biopsies to detect minimal residual disease (MRD). A positive MRD test may guide the use of adjuvant therapies like immunotherapy, aiming to cure microscopic cancer cells before they become clinically detectable.

5. Avoid Overtreatment via Repeat Testing

Explore the potential for serial liquid biopsy testing in patients at risk of recurrence to guide adjuvant therapy decisions. This approach could help avoid unnecessary treatment and its associated toxicities in patients who are genuinely cured, by only initiating therapy when a positive signal of residual cancer is detected.

6. Evaluate New Screening Tests Critically

When new cancer screening tests are introduced, compare their sensitivity and specificity against existing, proven screening methods. If a new test is not as good as established ones, carefully consider its utility, focusing on practical advantages like increased access or convenience rather than assuming superiority based on novelty.

7. Be Aware of Lung Cancer Risk

Understand that lung cancer is the number one cause of cancer death for both men and women, and 15% of those who die from it have never smoked. This awareness is crucial as anyone with lungs can develop the disease due to various risk factors beyond smoking.

8. Follow Data in Scientific Research

As a scientist, prioritize following the objective data generated by experiments, allowing it to guide your research direction rather than rigidly adhering to a preconceived area of focus. This adaptive approach can lead to unexpected and significant discoveries.

9. Ground Research in Clinical Needs

For physician-scientists, initiate research projects by identifying unmet clinical needs or suboptimal practices in patient care. This ensures that scientific endeavors are directly relevant and potentially impactful for improving diagnostic or treatment outcomes.

10. Leverage Startup Funds for Research

For new faculty members or researchers, strategically use initial startup funds provided by your department or university to pursue high-risk, potentially transformative research projects. This funding can kickstart work that may not yet qualify for traditional grant funding.

11. Recognize Research Reproducibility Challenges

Be aware that scientific findings, particularly in early-stage diagnostic development, may not always be easily reproducible or as effective as initially reported. When evaluating scientific literature, seek corroboration from multiple independent studies using diverse methods to identify robust findings.

12. Seek Professional Medical Advice

Always seek professional medical advice, diagnosis, or treatment from your healthcare professionals for any medical conditions. Do not disregard or delay obtaining medical advice based on information from this podcast or other general sources.

If you push on one, the other, you know, usually one goes up, most of the time the other goes down. Unless you really have a dramatic new insight or a new advance where you can just increase sensitivity without also hurting specificity.

Max Diehn

I think it's important for patients to understand the non-linearity of this. So you have everything shy of a billion cells is, by conventional detection methods, undetectable.

Peter Attia

In order for a screening test to be useful, it has to catch a significant fraction of the first type of stage one. The stage one that doesn't have micrometastases, which the surgeon can cure.

Max Diehn

Less tumor burden, less heterogeneity equals better outcomes. It's an axiom. That's absolutely true.

Peter Attia

Early detection absolutely matters. And you have to look no further than that simple point, which has, I've never seen it refuted.

Peter Attia

Guiding Adjuvant Therapy based on Minimal Residual Disease (MRD) Detection

Max Diehn
  1. Patient undergoes standard treatment (radiation or surgery) for early-stage lung cancer.
  2. If standard-of-care chemotherapy is indicated for their stage, they receive it.
  3. After completing all standard treatments, a blood test is performed to detect minimal residual disease (MRD) using a sensitive liquid biopsy.
  4. If the liquid biopsy is positive (detects ctDNA), the patient receives immunotherapy, even without visible cancer on scans.

Lung Cancer Screening using Mutation-Based Cell-Free DNA (Lung CLIP)

Max Diehn
  1. Sequence cell-free DNA (cfDNA) from the patient's blood plasma.
  2. Sequence leukocyte DNA (from white blood cells) from the same patient.
  3. Subtract mutations found in the leukocyte DNA (due to clonal hematopoiesis) from the cfDNA mutations to remove non-tumor signals.
  4. Apply a machine learning algorithm that analyzes remaining plasma mutations, considering factors like cfDNA molecule length, gene location of mutations, and whether mutations are associated with smoking.
  5. The model outputs a probability indicating the likelihood of lung cancer.
1 centimeter in diameter (or 8 millimeters)
Minimum tumor size detectable by traditional imaging (e.g., CT scan) Corresponds to approximately 1 billion cancer cells.
15%
Incidence of lung cancer in non-smokers Percentage of people who die of lung cancer who have never smoked.
Approximately 20%
Relative risk reduction in lung cancer deaths with low-dose CT screening Demonstrated in the National Lung Screening Trial compared to chest X-ray.
10 to 20 mL
Amount of blood typically collected for liquid biopsy tests Yields approximately 10 mL of plasma.
1 to 5 nanograms per milliliter
Concentration of cell-free DNA in healthy individuals' plasma Leading to 10 to 50 nanograms of DNA from 10 mL of plasma.
Approximately 160-170 bases
Typical length of cell-free DNA molecules Corresponds to the length of DNA wrapped around core histones.
Up to 1%
Fraction of circulating tumor DNA in advanced cancer patients Proportion of total cell-free DNA derived from cancer cells.
Less than 0.01% or 0.001%
Fraction of circulating tumor DNA in early-stage cancer patients Proportion of total cell-free DNA derived from cancer cells.
100 times
Sensitivity improvement of Max Diehn's third-generation mutation-based method Achieving a detection limit of one part in a million (0.0001%) for known mutations.
Approximately 0.1% (one in a thousand)
Estimated sensitivity of methylation-based assays for pan-cancer screening Inferior to mutation-based approaches for known mutations.
5% or less
Sensitivity of GRAIL test for stage one lung cancer Based on presented data for pan-cancer screening using methylation-based methods.
About 20%
Sensitivity of Diehn's mutation-based Lung CLIP method for stage one lung cancer Achieved in a validation cohort, with a specificity of 98-99%.