Table 3.
Hazard ratios for malignancy development within one year of mHLA-DRd quantification.
| 1 year cancer risk | Disease specific (CPH) | Competing Risk | |||
|---|---|---|---|---|---|
| Dependent: | All | HR (Univariate) | HR (Multivariate) | HR (Fine + Gray Multivariate) | |
| Age at enrolment (years) | Mean (SD) | 62.6 (10.8) | 1.08 (1.04-1.12, p<0.001) | 1.07 (1.03-1.12, p<0.001) | 1.07 (1.04-1.11, p<0.001) |
| Cumulative duration of immunosuppression (months) | Mean (SD) | 244.0 (103.2) | 1.00 (1.00-1.00, p=0.918) | 1.00 (0.99-1.00, p=0.434) | 1.00 (0.99-1.00, p=0.400) |
| Previous malignancy at enrolment | No | 119 (88.1) | |||
| Yes | 16 (11.9) | 0.55 (0.13-2.31, p=0.414) | 1.31 (0.29-5.88, p=0.721) | 1.33 (0.39-4.51, p=0.650) | |
| Taking prednisolone at recruitment | No | 89 (65.9) | |||
| Yes | 46 (34.1) | 0.73 (0.32-1.65, p=0.445) | 0.37 (0.15-0.89, p=0.027) | 0.37 (0.16-0.84, p=0.018) | |
| Chronic UV exposure at enrolment | No | 54 (40.0) | |||
| Yes | 81 (60.0) | 2.57 (1.04-6.35, p=0.040) | 2.19 (0.88-5.47, p=0.092) | 2.27 (0.86-5.94, p=0.096) | |
| Previous SCC | No | 77 (57.0) | |||
| Yes | 58 (43.0) | 4.80 (2.04-11.31, p<0.001) | 4.86 (1.95-12.07, p=0.001) | 4.84 (1.70-13.81, p=0.003) | |
| Enrolment mHLA-DRd | Low | 65 (48.1) | |||
| High | 70 (51.9) | 0.41 (0.18-0.90, p=0.027) | 0.29 (0.12-0.69, p=0.005) | 0.29 (0.12-0.70, p=0.006) | |
All factors analysed for univariate significant were included in multivariate models as a pre-specified analysis. Age at recruitment and age at last transplant demonstrated strong correlation, as did number of immunosuppressive agents and the use of prednisolone at recruitment, and so only age and use of prednisolone at recruitment were used in the multivariate models. Two models were generated: one using Cox Proportionate Hazards modelling and a second using competing-risk modelling, using graft loss and death as competing risks.