Han Liu 1, Lena Friberg 1
1 Department of Pharmacy, Uppsala University (Uppsala, Sweden)
Objectives
Biological sex contributes to variability in sunitinib pharmacokinetics, but the implications for tolerability, treatment modifications, and patient-reported side-effect burden remain insufficiently quantified. Using pooled metastatic renal cell carcinoma (mRCC) trial data, we aimed to (i) characterize sex differences in tolerability (dose reductions, adverse event [AE]–driven discontinuation, and patient-reported bother from side effects), (ii) quantify the extent to which observed sex differences are mediated by systemic exposure to sunitinib and its active metabolite SU12662, and (iii) explore sex-informed dosing strategies to improve benefit–risk.
Methods
We analyzed individual patient data from four clinical trials accessed via Vivli, restricted to patients initiating sunitinib at 50 mg once daily on the 4-weeks-on/2-weeks-off (4/2) regimen. A population pharmacokinetic (PK) model (two-compartment, first-order absorption/elimination) described sunitinib and SU12662, with inter-occasion variability added to enable cycle-specific steady-state exposure (AUCss). Total-drug exposure was defined as combined AUCss (sunitinib + SU12662), reflecting equipotency. Associations between exposure and clinician-driven treatment modifications were quantified in patients with PK data (n=294) using a multistate model comprising: initial dose, first dose reduction, second dose reduction, AE-driven discontinuation, and progression/other discontinuation. Exposure in cycle n predicted transitions in cycle n+1. Patient-reported bother from side effects was evaluated using FACT-G GP5 (5-point Likert scale) in patients with ≥1 post-baseline GP5 assessment (n=506; 3,751 observations) using a proportional-odds model, with exposure in the same cycle predicting GP5 at day 28. A dropout model accounted for informative missingness due to dose modification/discontinuation. Finally, simulations in a virtual cohort evaluated sex-informed starting doses (37.5 mg 4/2 in females) and an exposure-guided therapeutic drug monitoring (TDM) rule using combined trough concentration at day 14 (>100 ng/mL triggers −12.5 mg from day 15).
Results
Females exhibited higher systemic exposure than males. In the PK subset, median cycle-1 AUCss was higher in females versus males for sunitinib (1.88 vs 1.46 µg·h/mL) and SU12662 (1.10 vs 0.61 µg·h/mL), yielding a 40.8% higher combined AUCss (2.97 vs 2.11 µg·h/mL). In the multistate model, higher AUCss increased hazards of dose reduction and AE-driven discontinuation. A 1.0 µg·h/mL increase in combined AUCss corresponded to a 1.94-fold higher hazard for dose-reduction transitions and a 2.05-fold higher hazard for AE-driven discontinuation. In univariate analyses, females had poorer tolerability (more dose reductions and AE discontinuations), but after accounting for exposure, sex was no longer a significant predictor of treatment-modification transitions, while discontinuation due to progression/other reasons was similar between sexes.
In the PRO model, higher AUCss increased the probability of reporting worse GP5 (“bother from side effects”) with proportional effects across categories. After adjusting for exposure, sex-specific thresholds improved model fit: at comparable AUCss, females remained more likely to report higher bother levels (scores 3–5) than males, indicating that exposure explained part—but not all—of the sex difference in perceived side-effect burden.
Simulations suggested that a 37.5 mg 4/2 starting dose in females achieves concentration profiles comparable to those in males receiving 50 mg 4/2; at day 27, the median trough was 71.0 ng/mL in females at 37.5 mg versus 74.3 ng/mL in males at 50 mg. Under the TDM rule, dose reductions were triggered in 41% of females starting at 50 mg, 17% at 37.5 mg, and 18% of males beginning at 50 mg.
Conclusions
In sunitinib-treated mRCC, females have higher exposure to sunitinib + SU12662, which mediates observed sex differences in clinician-driven tolerability outcomes (dose reductions and AE-driven discontinuation). Exposure also predicts patient-reported side-effect bother, although residual sex differences persist at similar exposure. Sex-informed initiation (e.g., 37.5 mg 4/2 in females) and/or early exposure-guided down-titration may reduce toxicity burden without expected loss of efficacy, supporting precision dosing strategies that incorporate both exposure and patient-reported outcomes.
Reference: PAGE 34 (2026) Abstr 11955 [www.page-meeting.org/?abstract=11955]
Poster: Clinical Applications