Getting precise over multiple myeloma

Multiple myeloma is an incurable malignancy of immunoglobulin-secreting plasma cells.

Plasma cells are the workhorse B cells of the immune system, whose typical job is to wait patiently, poised to produce vast quantities of antibodies when required.

When they become cancerous, the consequences of this uncontrolled antibody production can be catastrophic. It can lead to irreversible renal failure – as the ‘sticky’ immunoglobulin light chains poison glomeruli in a range of ways – and a range of plasma hyperviscosity symptoms.

Alongside this, myeloma also drives a highly inflammatory state in its resident bone marrow niche, resulting in fatigue, haematopoietic and immune system dysfunction and dysregulation of bone turnover. This in turn produces the classic lytic bone lesions of myeloma, with often horrendous consequences for the integrity of the spine and central skeleton. Add this to the uncontrolled protein production, and you get the miserable constellation of symptoms that characterise the often-delayed presentation of this cancer.

Myeloma has a UK incidence of approximately 6,000 new cases/year, but its precursor condition smouldering myeloma affects 1 in 200 people over 40. 50% of these will progress to the active cancerous disease over a 5-year period. It is characterised by repeated disease relapses, near-universal drug resistance acquisition over time, and an increasingly complex and unstable genome. Median survival is around six to seven years but highly variable, depending on many factors including cancer genetics.

Despite extremely high-cost of treatment – 3-4 drug therapy combinations are required at each sequential therapy line – genetically defined ‘high-risk’ cases retain dismal prognosis, so precision-targeted therapies addressing high-risk myeloma biology are urgently needed.

Genetic clues

Healthy plasma cells are required to survive life-long and need to remain dormant but alive – unless antigen challenge arises. Their malignant counterpart myeloma cells escape this quiescence. This results from a complex assortment of proliferation-favouring, cancer-driving genetic hits acquired over decades. The key genetic hits that make a myeloma are largely structural, chromosome-level changes rather than mutational.

Unlike many other blood cancers, myeloma genomes can have a complexity approaching that of solid tumours. They particularly feature recurrent ‘first hit’ translocations of oncogenes which become placed under the transcriptional control of highly active immunoglobulin gene promotors, and recurrently selected whole, arm-length or smaller chromosome copy number alterations (CNAs).

The sub-clonal heterogeneity, genomic complexity, and relative ease of obtaining tumour material to study myeloma (repeat bone marrow biopsies are taken at recurrent relapses, heralded by rising clonal immunoglobulins tracked by blood testing) make it an excellent model for studying the evolution of a cancer genome over time.

As it develops from precursor states, the structure of its complexity is shaped by sequential multi-drug therapy exposures that drive the competition of genetically distinct subclones best suited to survive the different drug combinations. This evolution eventually renders it incurable.

Despite all this, therapeutic approaches are not currently precision-targeted, resulting in significant toxicity, variable responses and the heavy economic burden of long term, multi-drug regimes. This gap between an elusive potential and current reality continues to frustrate patients, clinicians, and scientists alike.

Why copy number alterations?

Recurrent CNAs are a feature of chromosomally unstable cancers. Myeloma carries fewer of them than many solid tumours, but there are some specific highly recurrent copy number aberrations in myeloma that carry prognostic implications.

Reproducible selection of particular CNAs in a cancer indicates that altered expression of a gene, or genes, it harbours favours progression or therapeutic resistance. Understanding the dysregulated biological processes driven by CNAs that are selected during cancer development is key here, as that may reveal targetable vulnerabilities – even if the target is not a gene from that CNA region itself.

However, unless it is the primary oncogenic founding event, targeting a single genetic variant always runs the risk of merely selecting an alternative clone that does not harbour the event, and therefore drives relapse.

Possibly more appealing is to hunt the underlying mechanism for the genetic instability that causes the CNA (or any other cancer hallmark) to occur. It can prove very valuable to ask why they arose in the first place and examine if the cancer cells are addicted to any particular cellular process that enables them to cope with the instability-driving mechanism.

Here, the field has much to learn from paradigm-shifting approaches in personalised medicine that are now mainstays in the treatment of some solid tumours. Classic examples here being the vulnerability of homologous recombination-deficient cancers to PARP inhibitors, or immune checkpoint blockade in MSI-high colorectal cancers.

In Ross Chapman’s genome integrity group in Oxford, we have been building on the discovery of just such a synthetic lethal interaction. He found the genome instability in a group of high-risk breast cancers characterised by 17q amplification is driven by high protein levels of the 17q-resident-gene TRIM37. This inhibits the production of normal levels of a matrix of proteins called the pericentriolar material (PCM), which together with centrosomes form microtubule organising centres, required for mitotic spindle organisation.

Whereas most other cancers can undergo mitosis without centrosomes to organise their spindle, these 17q amplified-cancers without adequate PCM are entirely dependent on their centrosomes alone to be able to complete mitosis. They are therefore hypersensitive to, and killed by, drugs that inhibit the replication and production of new centrosomes during cell division.

Synthetic lethality and myeloma?

Myeloma is not a cancer with sweeping signatures of a particular type of DNA repair deficiency, but we do have clues.

As with most B cell malignancies, primary oncogenic translocations result from errors in scheduled immunoglobulin gene rearrangement during B cell development. Copy number signatures associated with higher number and complexity of CNA events predict for chromothripsis (a chromosome shattering event, seen in 24% of myelomas) and poorer outcomes.

The mechanism by which premalignant myeloma clones slowly accumulate their CNAs during the many years over which they emerge, is not yet understood. However, it is clear that the highest risk myelomas show higher CNA complexity, the presence of chromothripsis events, and gene expression signatures suggestive of abnormal fidelity during cell division.

We have become interested in pinpointing abnormal mitotic behaviour in the highest risk myelomas – behaviours that might explain both the genetic events that accumulate, but also lead us to abnormal behaviours that we might exploit as cancer-specific weaknesses in new therapies.

If we can pinpoint the underlying cause of the mitotic error accumulation that we see in high-risk myeloma, we may find that there are essential mechanisms on which the myeloma cell is abnormally dependant, even compared to other cancers.

Such a dependence may enable a myeloma cell to complete a cell division – despite accompanying errors – and could just be our next target.