Novel Prognostic Implications of PAQR3 Expression in Breast Cancer after Curative Resection and Systemic Therapy

Background: This study aimed to explore the impact of progestin and adipoQ receptor 3 (PAQR3) on the outcomes for breast cancer patients who received curative resection and systemic therapy. Methods: The online databases consisting of 2,352 breast cancer patients were used to explore the association between PAQR3 expression and clinicopathological features. Univariate and multivariate survival analysis were performed to identify the survival predictive role of PAQR3. The internal validation strategies were applied. Results: PAQR3 expression was inversely correlated with estrogen receptor (ER) expression (P<0.0001), progesterone receptor (PR) expression (P<0.0001) and tumor grade (P<0.0001). Higher PAQR3 expression independently predicted shorter overall survival (OS) (hazard ratio (HR)=1.20, 95% condence interval (CI)=1.06-1.37, P=0.0055) and recurrence free survival (RFS) (HR=1.28, 95%CI=1.04-1.45, P=0.0170) in breast cancer patients. In ER+ breast cancer and PR+ breast cancer, PAQR3 was associated with shorter OS (ER+: HR=1.28, 95%CI=1.11-1.48, P=0.0006; PR+: HR=1.30, 95%CI=1.09-1.56, P=0.0037) and RFS (ER+: HR=1.46, 95%CI=1.19-1.77, P=0.0002; PR+: HR=1.48, 95%CI=1.16-1.89, P=0.0016), but no such prognostic values of PAQR3 expression were found in ER- breast cancer and PR- breast cancer. Conclusions: PAQR3 serves as an independent outcomes predictor for breast cancer.


Introduction
Breast cancer is the most frequently diagnosed cancer and the leading cause of cancer death among females worldwide [1]. There is an imperative need for the identi cation of molecular prognostic markers for breast cancer progression.
Progestin and adipoQ receptor (PAQR) superfamily are well known to be activated by the steroid hormones. It could be divided into three subfamilies: adiponectin receptor like proteins, membrane progestin receptor (mPR) containing subfamily, and pore channel forming hemolysins [2,3]. mPR is plasma membrane receptor involved in transducing nongenomic progesterone signaling [4]. mPR structure is predicted including a seven hydrophobic membrane spanning domains, a feature characteristic of GPCRs [5]. In breast cancer, mPR mediates progestin-induced cell survival [6] and progesterone-generated cancer stem cells [7], in a nuclear progesterone receptor independent manner.
Romero et al. found that the expression of mPR gamma was elevated in endometrioid and clear cell carcinomas, suggesting a potential role of the mPR in the pathogenesis of tumor [5].
Similar to mPR construction, PAQR3, a seven-transmembrane protein localized to the Golgi apparatus, is also considered regulating GPCR signaling pathway [8]. As previously reported, PAQR3 served as an inhibitor of the ERK activation and suppressor in tumors frequently harboring Ras/Raf mutation, such as melanoma [9], colorectal carcinoma [10], skin cancer [11], gastric cancer [12]. Actually, PAQR3 not only exists as the ERK phosphorylation inhibitor, it is also known as a progestin receptor. Therefore, it may Page 3/25 have a different role in progestin dependent cancers [13][14][15][16]. In this study, we aimed to demonstrate the prognostic value of PAQR3 on the outcomes for breast cancer patients who received curative resection and systemic therapy.

Materials And Methods
Public online data The data used in this study was downloaded from GOBO, TCGA and R2 website. The patients were divided into the higher and lower expression of PAQR3 at a speci c cut off value (6.56).

Statistics
The paired sample t-test was used to evaluate the difference of PAQR3 expression between tumor and paired normal tissues. The associations between PAQR3 expression and clinicopathological features were evaluated using the χ2 test. Survival probabilities were calculated by the Kaplan-Meier method, and the log-rank test was used to evaluate statistical difference. The Cox proportional hazards model was used for the multivariate analysis to identify the independent prognostic factors for patients' outcomes.
All statistical analyses were conducted using the R program. All tests were two sided, and a P < 0.0500 was considered signi cant.

PAQR3 expression was elevated in human breast cancer
We compared the PAQR3 expression between breast cancer and paired normal tissues. A decrease in ER and/or PR positive (ER/PR+) tumor tissues was found compared with the paired normal tissues (P < 0.0500), whereas no such difference was identi ed in ER and PR negative (ER/PR-) breast cancer (P > 0.0500) (Fig. 1A). With an online dataset (http://co.bmc.lu.se/gobo) [17] , we found PAQR3 expression was negatively associated with ER expression (P < 0.0001), whereas positively associated with tumor grade (P < 0.0001) (Fig. 1B).

Paqr3 Expression Was Associated With Breast Cancer Clinicopathological Features
The associations between PAQR3 expression and clinicopathological features were further explored in 1,873 cases of breast cancer (Cohort A). The demographics and baseline clinicopathological characteristics were listed (Table 1). Tumors with higher levels of PAQR3 were more likely to occur in the patients aged less than 70 years (P < 0.0001) and those at the period of premenopause (P < 0.0001). In   Another cohort of 479 breast cancer patients were applied to validate the above ndings (Cohort B). The age and baseline characteristics were described (Table S1). Similarly, PAQR3 was associated with the higher tumor grade (P = 0.0080), ER loss (P < 0.0001) and PR loss (P = 0.0020) ( Table 3).  overexpression, the higher PAQR3 expression remained serving as an independent predictor for shorter OS in the training set (HR = 1.24, P = 0.0206), and was likely to predict the OS in the validation set (HR = 1.18, P = 0.0884) (Table S2). However, we did not nd that the higher PAQR3 expression predicted shorter RFS in training and validation sets.

Paqr3 Better Predicted The Outcomes In Er/pr + breast Cancer
As ER/PR status was associated with PAQR3 expression, we strati ed the 1,873 patients according to the ER and PR status, and then studied if PAQR3 expression predicted the outcomes in ER + and PR + breast cancer, respectively.
Internally validating the predictive role of PAQR3 in ER/PR + breast cancer We veri ed the predictive role of PAQR3 expression for breast cancer patients' outcomes according to the ER and PR status. The previously assigned training (N = 936) and validation (N = 937) sets were utilized here.

Discussion
Previous studies have demonstrated the involvement of PAQR3 in various cancers, but its role in breast cancer, whose progression mostly relies on sexual hormone regulation, remains unclear. Chen et al. rstly described that PAQR3 could trap B-Raf to the Glogi apparatus thus inhibit ERK activation [9]. Nevertheless, all previous studies just concentrated on PAQR3's role in ERK inactivation, which they thought contributing to its tumor suppressing role [18]. None considered PAQR3 could also be a potential mediator for the rapid nongenomic hormone actions [4].
It is well known that progestin could increase breast cancer incidence and promote tumor progression through the combination with mPR [19,20]. Moreover, Dressing et al. proposed a potential involvement of mPR in the progression of breast cancer, since the intracellular signaling initiated by the mPR is broadly similar to that induced in breast cancer growth [21]. mPR was also found to mediate tumor cell survival [6] and cancer stem cells generation [7]. The mPR expression could be regulated by progesterone, one study found an increase in progesterone was shown to coincide with a reduced level of mPR levels [2]. In our study, we found PAQR3 levels were higher in ER-, PR-or poorly differentiated breast cancer, which are well known as the major risk factors for poor prognosis in breast cancer. These suggested that PAQR3 expression distribution in breast cancer differed from other cancers and PAQR3 may possess a promotion role in breast cancer progression.
In this study, we demonstrated for the rst time that higher PAQR3 expression was associated with poor outcomes in breast cancer, especially in ER + breast cancer. Giulianelli et al. found that progestin-induced breast cancer proliferation was dependent on ER, which possibly explained why PAQR3 better predicted the prognosis in ER + breast cancer [15]. The PAQR3's predictive role found in this study was quite different from its role in other cancers, the following points may explain the divergence: 1) Ras/Raf mutation is not common in breast cancer, compared with melanoma and colorectal cancer which possess a high frequency of Ras/Raf mutation [13,22]; This also means the ERK activation was obvious for breast cancer, thus the ERK inactivation role of PAQR3 was not important in breast cancer; 2) Breast cancer is different from other cancers, it could be regulated by sexual hormones such as progestin, which was recognized as the ligand of PAQR3; 3) No such a large cohort of patients was used for prognosis analysis. Therefore, within the ndings we mentioned above, we concluded that PAQR3 may possess a promotion role in tumor progression but not occurrence, and its expression might be downregulated by ER and/or PR expression, this could explain why PAQR3 expression decrease could be only found in ER/PR + tumor tissues compared with paired normal tissues.
In summary, to our knowledge, this is the rst large sample study of PAQR3's predictive role in breast cancer patients' outcomes. The ndings in this study may help understand PAQR3's role in breast cancer, but still requires a further study.

Declarations
Ethics approval and consent to participate None declared.

Consent for publication
Not applicable.
Competing interest None declared.