Fig 2: Correlation between the combined panels of serum AAb levels and OST of CMT dogs. CMT dogs (n = 90) were dichotomized by a combination of (A) TYMS-AAbHigh and AGR2-AAbLow (denoted T-AAbHigh/A-AAbLow), (B) IGFBP5-AAbHigh and AGR2-AAbLow (denoted I-AAbHigh/A-AAbLow), (C) TYMS-AAbHigh and IGFBP5-AAbHigh groups (denoted T-AAbHigh/I-AAbLow), and (D) TYMS-AAbHigh, IGFBP5-AAbHigh, and AGR2-AAbLow (denoted T-AAbHigh/I-AAbHigh/A-AAbLow), respectively. OST of CMT dogs was compared between the dichotomous groups by using Kaplan-Meier analysis. Statistical significance was determined by the log-ranked test.

Fig 3: Correlation between the combined panels of serum AAb levels and OS of dogs with non-metastasized MMT. Dogs with non-metastatic MMT (n = 59 in stage I, II, or III) were dichotomized by a combination of (A) TYMS-AAbHigh and AGR2-AAbLow (denoted T-AAbHigh/A-AAbLow), (B) IGFBP5-AAbHigh and AGR2-AAbLow (denoted I-AAbHigh/A-AAbLow), and (C) TYMS-AAbHigh, IGFBP5-AAbHigh, and AGR2-AAbLow (denoted T-AAbHigh/I-AAbHigh/A-AAbLow), respectively. OST of non-metastasized MMT dogs was compared between the dichotomous groups by using Kaplan-Meier analysis. Statistical significance was determined by the log-ranked test.

Fig 4: Correlation between the combined panels of serum AAb levels and OS of MMT dogs. Dogs with MMT (n = 70) were dichotomized by a combination of (A) TYMS-AAbHigh and AGR2-AAbLow (denoted T-AAbHigh/A-AAbLow), (B) IGFBP5-AAbHigh and AGR2-AAbLow (denoted I-AAbHigh/A-AAbLow), (C) TYMS-AAbHigh and IGFBP5-AAbHigh (denoted T-AAbHigh/I-AAbLow), and (D) TYMS-AAbHigh, IGFBP5-AAbHigh, and AGR2-AAbLow (denoted T-AAbHigh/I-AAbHigh/A-AAbLow), respectively. OST of MMT dogs was compared between the dichotomous groups by using Kaplan-Meier analysis. Statistical significance was determined by the log-ranked test.

Fig 5: Elevated protein levels of AGR2 in canine MMT tissues compared with paired normal mammary tissues. Tissue homogenates (20 µg proteins each sample) of MMT (denoted M), paired with benign mammary tumor (denoted B) and normal mammary gland tissues (denoted N) collected from the same patient, were analyzed by immunoblotting using an antibody specific to AGR2. GAPDH was used as an internal control. Representative results of nine cases (No. 1 to No. 9) were shown in (A). AGR2 protein levels in individual samples were determined by calculating the AGR2-to-GAPDH ratio. (B) Comparison of AGR2 protein levels between tissue types. Statistical significance was determined by Mann–Whitney U test. (C) Pairwise comparison of AGR2 protein levels between MMT and paired normal mammary gland tissues of the same patient. Lines represent individual patients. Statistical significance was determined by two-tailed paired t-test. ** p < 0.01; *** p < 0.001.