Bayesian Pathway Analysis for Complex Interactions

Baurley, James W.; Kjærsgaard, Anders; Zwick, Michael E.; Cronin‐Fenton, Deirdre; Collin, Lindsay J.; Damkier, Per; Hamilton-Dutoit, Stephen; Lash, Timothy L.; Ahern, Thomas P.

doi:10.1093/aje/kwaa130

Cited by 3 publications

(3 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such research could help identify women at risk of poor recovery after taxane-based chemotherapy. As we only examined the associations of single SNPs, future studies should include Bayesian pathway analysis considering the entire complex metabolic pathway of docetaxel [ 18 , 55 ].…”

Section: Discussionmentioning

confidence: 99%

The impact of single nucleotide polymorphisms on return-to-work after taxane-based chemotherapy in breast cancer

Hjorth

Damkier

Stage

et al. 2023

Cancer Chemother Pharmacol

Self Cite

View full text Add to dashboard Cite

Purpose Breast cancer treatment is associated with adverse effects, which may delay return-to-work. Single nucleotide polymorphisms (SNPs) may influence the risk and severity of treatment toxicities, which in turn could delay return-to-work. We examined the association of 26 SNPs with return-to-work in premenopausal women with breast cancer. Methods Using Danish registries, we identified premenopausal women diagnosed with non-distant metastatic breast cancer during 2007‒2011, assigned adjuvant combination chemotherapy including cyclophosphamide and docetaxel. We genotyped 26 SNPs in 20 genes (ABCB1, ABCC2, ABCG2, CYP1A1, CYP1B1, CYP3A, CYP3A4, CYP3A5, GSTP1, SLCO1B1, SLCO1B3, ARHGEF10, EPHA4, EPHA5, EPHA6, EPHA8, ERCC1, ERCC2, FGD4 and TRPV1) using TaqMan assays. We computed the cumulative incidence of return-to-work (defined as 4 consecutive weeks of work) up to 10 years after surgery, treating death and retirement as competing events and fitted cause-specific Cox regression models to estimate crude hazard ratios (HRs) and 95% confidence intervals (CIs) of return-to-work. We also examined stable labor market attachment (defined as 12 consecutive weeks of work). Results We included 1,964 women. No associations were found for 25 SNPs. The cumulative incidence of return-to-work varied by CYP3A5 rs776746 genotype. From 6 months to 10 years after surgery, return-to-work increased from 25 to 94% in wildtypes (n = 1600), from 17 to 94% in heterozygotes (n = 249), and from 7 to 82% in homozygotes (n = 15). The HR showed delayed return-to-work in CYP3A5 rs776746 homozygotes throughout follow-up (0.48, 95% CI 0.26, 0.86), compared with wildtypes. Estimates were similar for stable labor market attachment. Conclusion Overall, the SNPs examined in the study did not influence return-to-work or stable labor market attachment after breast cancer in premenopausal women. Our findings did suggest that the outcomes were delayed in homozygote carriers of CYP3A5 rs776746, though the number of homozygotes was low.

show abstract

Section: Discussionmentioning

confidence: 99%

The impact of single nucleotide polymorphisms on return-to-work after taxane-based chemotherapy in breast cancer

Hjorth

Damkier

Stage

et al. 2023

Cancer Chemother Pharmacol

Self Cite

View full text Add to dashboard Cite

show abstract

“…Findings from this study merit further investigation. They will be used in future work using multiple pathway analysis [ 23 , 57 ] to capture the net effect of these SNPs. Also, future clinical trials are needed to elucidate whether genomic testing could guide dosing of taxane-based therapy to reduce inter-individual variability in effectiveness.…”

Section: Discussionmentioning

confidence: 99%

Single-nucleotide polymorphisms and the effectiveness of taxane-based chemotherapy in premenopausal breast cancer: a population-based cohort study in Denmark

Hjorth

Damkier

Stage

et al. 2022

Breast Cancer Res Treat

Self Cite

View full text Add to dashboard Cite

Purpose Taxane-based chemotherapy is the primary treatment for premenopausal breast cancer. Although being inconsistent, research suggests that variant alleles alter pharmacokinetics through reduced function of OATP transporters (limiting hepatic uptake), CYP-450 enzymes (hampering drug metabolism), and ABC transporters (decreasing clearance). Reduced function of DNA repair enzymes may hamper effectiveness through dose-limiting toxicities. We investigated whether single-nucleotide polymorphisms (SNPs) were associated with breast cancer recurrence or mortality in premenopausal women diagnosed with breast cancer. Methods We conducted a population-based cohort study of premenopausal women diagnosed with non-distant metastatic breast cancer in Denmark during 2007‒2011, when guidelines recommended adjuvant combination chemotherapy (taxanes, anthracyclines, and cyclophosphamide). Using archived formalin-fixed paraffin-embedded primary tumor tissue, we genotyped 26 SNPs using TaqMan assays. Danish health registries provided data on breast cancer recurrence (through September 25, 2017) and death (through December 31, 2019). We fit Cox regression models to calculate crude hazard ratios (HRs) and 95% confidence intervals (CIs) for recurrence and mortality across genotypes. Results Among 2,262 women, 249 experienced recurrence (cumulative incidence: 13%) and 259 died (cumulative incidence: 16%) during follow-up (median 7.0 and 10.1 years, respectively). Mortality was increased in variant carriers of GSTP1 rs1138272 (HR: 1.30, 95% CI 0.95–1.78) and CYP3A rs10273424 (HR: 1.33, 95% CI 0.98–1.81). SLCO1B1 rs2306283 (encoding OATP1B1) variant carriers had decreased recurrence (HR: 0.82, 95% CI 0.64–1.07) and mortality (HR: 0.77, 95% CI 0.60–0.98). Conclusion Docetaxel effectiveness was influenced by SNPs in GSTP1, CYP3A, and SLCO1B1 in premenopausal women with non-distant metastatic breast cancer, likely related to altered docetaxel pharmacokinetics. These SNPs may help determine individual benefit from taxane-based chemotherapy.

show abstract

“…A recent approach, Algorithm for Learning Pathway Structures (ALPS), which uses a Monte Carlo scheme to update a pathway structure has shown promise for identifying complex interactions in large epidemiological datasets. [50] However, since ALPS focuses on parameter interpretation, its results differs from those of the CoOL approach, which identifies sizeable sub-groups who share exposures, which may have led to their increased risk of the outcome.…”

Section: Theoretical Comparison With Other Approachesmentioning

confidence: 99%

Causes of Outcome Learning: A causal inference-inspired machine learning approach to disentangling common combinations of potential causes of a health outcome

Rieckmann

Dworzyński

Arras

et al. 2020

Preprint

View full text Add to dashboard Cite

Nearly all diseases can be caused by different combinations of exposures. Yet, most epidemiological studies focus on the causal effect of a single exposure on an outcome. We present the Causes of Outcome Learning (CoOL) approach, which seeks to identify combinations of exposures (which can be interpreted causally if all causal assumptions are met) that could be responsible for an increased risk of a health outcome in population sub-groups. The approach allows for exposures acting alone and in synergy with others. It involves (a) a pre-computational phase that proposes a causal model; (b) a computational phase with three steps, namely (i) analytically fitting a non-negative additive model, (ii) decomposing risk contributions, and (iii) clustering individuals based on the risk contributions into sub-groups based on the predefined causal model; and (c) a post-computational phase on hypothesis development and validation by triangulation on new data before eventually updating the causal model. The computational phase uses a tailored neural network for the non-negative additive model and Layer-wise Relevance Propagation for the risk decomposition through this model. We demonstrate the approach on simulated and real-life data using the R package ‘CoOL’. The presentation is focused on binary exposures and outcomes but can be extended to other measurement types. This approach encourages and enables epidemiologists to identify combinations of pre-outcome exposures as potential causes of the health outcome of interest. Expanding our ability to discover complex causes could eventually result in more effective, targeted, and informed interventions prioritized for their public health impact.

show abstract

Bayesian Pathway Analysis for Complex Interactions

Cited by 3 publications

References 33 publications

The impact of single nucleotide polymorphisms on return-to-work after taxane-based chemotherapy in breast cancer

The impact of single nucleotide polymorphisms on return-to-work after taxane-based chemotherapy in breast cancer

Single-nucleotide polymorphisms and the effectiveness of taxane-based chemotherapy in premenopausal breast cancer: a population-based cohort study in Denmark

Causes of Outcome Learning: A causal inference-inspired machine learning approach to disentangling common combinations of potential causes of a health outcome

Contact Info

Product

Resources

About