Cancer progression and novel treatments
What are the most promising molecular targets for new and repurposed chemoprevention and therapeutic drugs? Our Cancer progression and novel treatments theme identifies molecular mechanisms in cancer survival, conducts economic evaluation of cancer screening, investigates drug repurposing using Mendelian randomization (MR), synthesises evidence for treatments and evaluates physical activity interventions to enhance cancer survival.
Highlights include:
Statins and ovarian cancer risk
We evaluated the effects of LDL-lowering drugs on epithelial ovarian cancer risk, finding evidence that inhibition of the target of statins (HMG-CoA reductase) significantly lowered the risk of ovarian cancer risk amongst sporadic cases and BRCA1/2 carriers. More information: doi.org/10.1001/jama.2020.0150
New insights into application of Mendelian randomization to cancer survival
There is uncertainty over whether Mendelian randomization (MR) can be used to identify treatments for prevention of mortality after a cancer diagnosis. We have addressed this uncertainty through a MR study of PD1 and PDL1 - proteins that regulate the immune system response to cancer; pharmaceutical inhibition of these proteins is approved for the treatment of several cancers.
We found that genetically proxied PD-1 lowering was associated with reduced risk of lung and ovarian cancer death (in the expected direction) and possibly melanoma (although with a wide confidence interval). These findings suggest that PD-1 inhibitors could be repurposed to treatment of ovarian cancer (which does not currently have any approved indications for PD1 or PDL1 inhibitors).
In contrast, we saw very little evidence for associations of genetically proxied PD-L1 lowering with breast, lung, melanoma, prostate or ovarian cancer death, with hazard ratios opposite to the expected direction or close to 1, despite the fact that PD-L1 inhibition is an approved treatment for several cancers (breast cancer, lung cancer and melanoma).
These findings could reflect low power or poor instrument validity (e.g. clinical benefit of PD-L1 inhibition depends on tumour expression, which might not be captured by germline variants). Overall, these findings identify a potential repurposing opportunity for ovarian cancer and provide valuable insights into current challenges of applying MR in cancer survival settings. These findings have been presented at the Bristol Mendelian randomization conference and will also be presented at the World Congress of Epidemiology in Cape Town. More information: https://bmjopen.bmj.com/content/14/2/e075981
Aromatase inhibitors could be repurposed for prevention of endometrial cancer
Aromatase inhibitors are approved for prevention of ER+ breast cancer in high-risk women. Through MR analyses, we found that genetically proxied aromatase inhibition reduced risk of ER+ breast cancer risk (a known effect) and endometrial cancer risk (not currently known). These results suggest that aromatase inhibitors could be repurposed for endometrial cancer prevention. These findings have been presented at the Bristol Mendelian randomization conference and will also be presented at the upcoming World Congress of Epidemiology in Cape Town.
Understanding the heritable basis of survival after a lung cancer diagnosis
To understand the heritable basis of survival after a lung cancer diagnosis, we are carrying out the largest meta-analysis of genome-wide association studies (GWAS) of lung cancer survival ever conducted. We plan to complete the meta-analysis by the end of 2024, with publication in 2025. The results of the study will provide insight into new potential targets for treatment of lung cancer after diagnosis.
Failure to get drugs to market is high, and treatment effectiveness is often determined by waiting to observe effects on cancer progression and survival. In this theme, we aim to address this issue by identifying high-confidence drug targets, and opportunities for drug repurposing, using Mendelian randomization (MR).
MR is a method that uses genetic variation to evaluate causal relationships; it has shown promise for correctly predicting outcomes of clinical trials. We are extending this method, which has been used primarily to establish effects on cancer incidence, to evaluate therapeutic effects on progression and survival after diagnosis.
We are also improving treatment decision-making through systematic biomarker discovery of disease prognosis and response to therapy. This is achieved within relevant peripheral and biopsy tissues using state-of-the-art molecular profiling and machine learning techniques.
