Genetics of Cancer 2: Advanced Cancer Genetics and Treatments

Cancer-critical genes Tumor Suppressors wild type

wild-type function is to prevent cancer development

1. stop cell division when inappropriate

2. prevent or fix mutations

3. cause programmed cell death

4. prevent motility

Cancer-critical genes Tumor Suppressors loss of function

loss-of function mutations (or expression changes) favor cancer development

1. methylation

2. loss of promoter/enhancer function

3. mutations in exons that effect function

proto-oncogene

wild-type function is to increase cell division or mobility, block cell death

oncogene

gain of function mutations (or expression changes) favor cancer development

1. overexpression

2. hyperactivation

3. constitutive activation

Techniques to understand the genetics of cancer 1

Cancer is a broad group of diseases. Multiple combinations of multiple genes can produce the same outcome in different tissues.

Techniques to understand the genetics of cancer 2

Some treatments target specific gene functions (e.g. Myc inhibitor, E2F inhibitor, DSB-inducing agent).

Techniques to understand the genetics of cancer 3

Many of these genes are broadly involved in cell division, therefore these treatments broadly target rapidly dividing cells (like cancer!), but also cells that normally divide rapidly (follicle cells, gut lining, red blood stem cells).

Techniques to understand the genetics of cancer 4

Understanding the specifics of one patient’s cancer can inform treatment.

• E.g. Tamoxifen – targets estrogen receptors.

Techniques to understand the genetics of cancer

• Cancer is a broad group of diseases. Multiple combinations of multiple genes can produce the same outcome in different tissues.

• Some treatments target specific gene functions (e.g. Myc inhibitor, E2F inhibitor, DSB-inducing agent).

• Many of these genes are broadly involved in cell division, therefore these treatments broadly target rapidly dividing cells (like cancer!), but also cells that normally divide rapidly (follicle cells, gut lining, red blood stem cells).

• Understanding the specifics of one patient’s cancer can inform treatment.

  • E.g. Tamoxifen – targets estrogen receptors.

Cancer heterogeneity

• Cancers vary based on tissue of origin

• Cancers can vary within a single tumor

TRACERx study:

Focus on Non-Small-Cell Lung Cancers

• Recruit study participants before they receive any treatment for their cancers

• Genetic analysis of tumor samples compared to genomes from non-cancerous tissue

• “Clonal” changes are present in every sample from the tumor

• “Subclonal” changes are present in a subset of tumor samples within the individual, not all

Observations from TRACERx study

1. high variability between individuals in total number of mutations ( up to ~3500)

2. some patients had mostly clonal mutations, others had most mutations subclonal

3. some patients had all clonal copy number variation, some had mostly subclonal, some zero

Tracking tumors over time shows evolution 1

Parallel evolution: Multiple lineages end up with alterations in same regions independently. This can be seen using SNP arrays

Tracking tumors over time shows evolution 2

early mutations shared by all tumor locations occurred earlier in the “trunk” part of this diagram

BCL11A is overexpressed in many cancers.

BCL11A is a transcriptional repressor that regulates some activators of tumor suppressors. 

Evolution-informed treatment approaches

• Targeting multiple common clonal events

  • would take multiple simultaneous mutations to resist the therapy

• Sequential treatments exploiting evolutionary bottlenecks

  • Drug #1 favors survival of a small number of cells

  • Drug #2 targets the most likely mutations that allowed survival 

Inherited SNPs might be

linked to rare alleles.

GWAS

Find SNPs associated with disease

1. SNP doesn’t necessarily reveal causative mutation. Can use EXPERIMENTAL data from other sources.

2. Linkage Disequilibrium (LD): Tells you regions that are linked to your SNP.

3. Cell specificity: In which cells are these loci generally active?

4. How does gene expression correlate with variants?

Which are the causal variants?

Fine-mapping

In which cell types do the variants act?

SNP enrichment

Which genes are regulated by the variants?

colocalization

Question 1: Why might certain SNPs overlap between multiple cancers?

Epistatic dependencies: Two different loci strongly drive cancer.

Enhancer regulation: These overlapping regions may represent chromatin regions that physically interact.

Question 2: NONE of the SNPs mapped in this diagram are located within exons. How might they be influencing cancer predisposition?

Linkage: The identified SNPs might not be causative, but are ASSOCIATED with causative alleles.

Regulation: The SNPs might alter gene regulation.

Your immune system attacks

“foreign” cells

• T-cells discriminate self from non-self cells via MHC display.

  • MHC is a protein that “displays” antigens for the T-cells.

• Due to the heavy mutational load of cancer cells, they often display THE WRONG antigens on the surface.

• These wrong signals are interpreted as “foreign” by T-cells, which initiate an immune response against the tumor cell.

• Knowing which antigens a tumor cell displays can guide therapy.

CAR – T Therapy:

Chimeric Antigen Receptors

• Engineer your own antigen receptor using part of an antibody (scFV = single chain variable fragment).

• Link this fragment to an intracellular domain that triggers T-cell response.

CAR – T Therapy vs sickle cell

• Like the sickle-cell therapy we discussed, this therapy requires genetically engineering your cells.

• KEY DIFFERENCES:

  • A virus is used to force the T-cells to produce the CAR protein. The cells are essentially “infected” with a virus that produces A LOT of this protein.

  • These T-cells DO NOT replace your native T-cells, they simply compliment them.

  • The patient is not necessarily genetically engineered, only a TINY number of cells (T-cells) are engineered. These cells eventually die.

CAR-T Challenges.

1. Antigen Escape.

   1. Cancer cells RAPIDLY evolve. Strong selection for cells that DO NOT display antigen.

2. Target antigen may also be produced by NORMAL cells.

3. CAR-T cells have trouble infiltrating thick tumors.

4. Cancer cells often evolve immunosuppressive properties (which is how they escape your immune system in the first place.

5. Cytokine Release Syndrome.

CAR-T Challenges: Antigen escape

• IDEALLY, you would sequence and monitor cancer tissue, and design CAR-T cells with updates.

• This is PROHIBITIVELY expensive and time consuming.

INSTEAD:

• Use a combination of antigen targets that are GENERALLY known to be displayed for a particular type of cancer. 

CAR-T Challenges: Antigen specificity

• Not all antigens are necessarily cancer-specific.

SOLUTION

• Many cancers produce rare/abnormal glycosylations.

• Design scFv to target these rare modifications.

CAR-T Challenges: Cytokine Release Syndrome 

• An immune response triggers the release of cytokines.

• These signal the immune system to recruit more macrophages and other immune cells to ramp up response.

• In some individuals, this escalates to unhealthy levels.

Summary

• Cancers differ with respect to which genes are mutated or have altered copy numbers

• Different regions of the same tumor can vary in their genetic makeup

• Cancer treatments can sometimes be customized to treat the specific changes in the cancer being treated.

• Some cancer treatments can leverage our own immune system to target cancer cells, specifically.