Results from several studies show that only obese, unfit subjects, but not obese, fit subjects, are at higher mortality risk than are normal-weight fit subjects. The aim of the study was two-fold: (1) to examine the differences in C-reactive protein levels across different metabolic phenotypes (healthy and unhealthy) of weight status and (2) ascertain whether high levels of cardiorespiratory fitness (CRF) attenuate the association of C-reactive protein and metabolic phenotypes of weight status. This was a pooled study, which included data from three cross-sectional projects (1706 youth (921 girls) aged 12–18 years). We used a Shuttle run test to assess CRF. Adolescents were classified into six metabolic phenotypes (healthy and unhealthy) of weight status (non-overweight, overweight and obese), based on age- and sex-specific cutoff points for triglycerides, systolic blood pressure, HDL-cholesterol, glucose and body mass index. High-sensitivity assays were used to obtain the C-reactive protein as inflammatory biomarker. After adjustment for potential confounders (age, sex, pubertal stage and country), the analysis of covariance (ANCOVA) shows that C-reactive protein is directly associated with metabolic phenotypes of weight status. Subjects with obesity, regardless of their metabolic profile, had higher levels of C-reactive protein Z-score. In addition, (after adjustments for potential confounders) a two-way ANCOVA showed that high levels of CRF attenuated the associations of C-reactive protein levels in metabolic healthy non-overweight and in adolescents with obesity. In conclusion, higher CRF levels may attenuate the detrimental association between obesity and C-reactive protein independently of metabolic phenotype. Findings from this study are important for prevention, clinical practice on issues associated with adiposity and metabolic disorders.