Predictive and Prognostic Brain Metastases Assessment in Luminal Breast Cancer Patients: FN14 and GRP94 from Diagnosis to Prophylaxis

FN14 has been implicated in many intracellular signaling pathways, and GRP94 is a well-known endoplasmic reticulum protein regulated by glucose. Recently, both have been associated with metastasis progression in breast cancer patients. We studied the usefulness of FN14 and GRP94 expression to stratify breast cancer patients according their risk of brain metastasis (BrM) progression. We analyzed FN14 and GRP94 by immunohistochemistry in a retrospective multicenter study using tissue microarrays from 208 patients with breast carcinomas, of whom 52 had developed BrM. Clinical and pathological characteristics and biomarkers expression in Luminal and non-Luminal patients were analyzed using a multivariate logistic regression model adjusted for covariates, and brain metastasis-free survival (BrMFS) was estimated using the Kaplan–Meier method and the Cox proportional hazards model. FN14 expression was associated with BrM progression mainly in Luminal breast cancer patients with a sensitivity (53.85%) and specificity (89.60%) similar to Her2 expression (46.15 and 89.84%, respectively). Moreover, the likelihood to develop BrM in FN14-positive Luminal carcinomas increased 36.70-fold (3.65–368.25, p = 0.002). Furthermore, the worst prognostic factor for BrMFS in patients with Luminal carcinomas was FN14 overexpression (HR = 8.25; 95% CI: 2.77–24.61; p = 0.00015). In these patients, GRP94 overexpression also increased the risk of BrM (HR = 3.58; 95% CI: 0.98–13.11; p = 0.054—Wald test). Therefore, FN14 expression in Luminal breast carcinomas is a predictive/prognostic biomarker of BrM, which combined with GRP94 predicts BrM progression in non-Luminal tumors 4.04-fold (1.19–8.22, p = 0.025), suggesting that both biomarkers are useful to stratify BrM risk at early diagnosis. We propose a new follow-up protocol for the early prevention of clinical BrM of breast cancer patients with BrM risk.

FN14 has been implicated in many intracellular signaling pathways, and GRP94 is a well-known endoplasmic reticulum protein regulated by glucose. Recently, both have been associated with metastasis progression in breast cancer patients. We studied the usefulness of FN14 and GRP94 expression to stratify breast cancer patients according their risk of brain metastasis (BrM) progression. We analyzed FN14 and GRP94 by immunohistochemistry in a retrospective multicenter study using tissue microarrays from 208 patients with breast carcinomas, of whom 52 had developed BrM. Clinical and pathological characteristics and biomarkers expression in Luminal and non-Luminal patients were analyzed using a multivariate logistic regression model adjusted for covariates, and brain metastasis-free survival (BrMFS) was estimated using the Kaplan-Meier method and the Cox proportional hazards model. FN14 expression was associated with BrM progression mainly in Luminal breast cancer patients with a sensitivity (53.85%) and specificity (89.60%) similar to Her2 expression (46.15 and 89.84%, respectively). Moreover, the likelihood to develop BrM in FN14-positive Luminal carcinomas increased 36.70-fold (3.65-368. 25, p = 0.002). Furthermore, the worst prognostic factor for BrMFS in patients with Luminal carcinomas was FN14 overexpression (HR = 8.25; 95% CI: 2.77-24.61; p = 0.00015). In these patients, GRP94 overexpression also increased the risk of BrM (HR = 3.58; 95% CI: 0.98-13.11; p = 0.054-Wald test). Therefore, FN14 expression in Luminal breast carcinomas is a predictive/prognostic biomarker of BrM, which combined with GRP94 predicts BrM progression in non-Luminal tumors 4.04-fold Prediction of BCBrM Frontiers in Oncology | www.frontiersin.org December 2017 | Volume 7 | Article 283 (1.19-8.22, p = 0.025), suggesting that both biomarkers are useful to stratify BrM risk at early diagnosis. We propose a new follow-up protocol for the early prevention of clinical BrM of breast cancer patients with BrM risk.
Keywords: biomarkers, brain metastasis, breast cancer, Fn14, grP94, prediction, prevention, prognosis inTrODUcTiOn Identification of molecular subtypes has enhanced our understanding of breast cancer biology (1), overcoming one of the main barriers to improving the progression, prognosis, and treatment of breast cancer, namely, its clinical and genetic heterogeneity. The gene expression patterns derived from cDNA microarrays of primary breast carcinomas have made it possible to correlate tumor characteristics with clinical outcome (2) and support the idea that breast tumor subtypes represent biologically distinct disease entities with different survival rates (3). The main recognized breast cancer subtypes are as follows: Luminal A, estrogen-receptor (ER) positive, Ki-67 < 14%, and normal expression of Her2; Luminal B, ER-positive, Ki-67 ≥ 14%, and normal expression of Her2; Luminal/Her2+, ER-positive and Her2 overexpression; Her2-enriched, ER-negative and Her2 overexpression; and triple negative (TN), ER-negative, progesterone receptor (PR) negative, and normal expression of Her2. One of the important differences between subtypes as regards clinical progression is that hormone receptor-positive tumors, such as Luminal A, have a better prognosis for survival compared with Her2 overexpression and TN subtypes (4,5) and the lowest risk of lymph node metastasis, whereas the Luminal-Her2+ subtype has the highest risk (6). Moreover, hormone receptor-positive subtypes such as Luminal A and Luminal B should be considered different oncologic entities sharing similarities when studying their pattern of response to therapy (7). Breast cancer molecular subtypes are used to stratify patients at increased risk of recurrence, who may benefit from more aggressive local treatment (8)(9)(10). For example, the Luminal/Her2+ and Her2-enriched subtypes are associated with a significantly higher rate of brain, lung, and liver metastases in comparison with the Luminal A subtype, whereas TN patients are associated with a higher rate of brain, lung and distant nodal metastases (11)(12)(13)(14). Despite improvements in diagnosis and novel adjuvant therapies, brain metastasis (BrM) is becoming a serious clinical problem, with a higher incidence in patients with histological grade (HG) 3, high Ki-67 expression (15), age younger than 50 years old (11,16), ER-negative and Her2-positive (11,17). Breast cancer subtypes also determine the prognosis and survival of a patient with BrM (18,19). Patients with Luminal tumors have a better survival rate than those with TN tumors (20), whereas those with the Her2-enriched subtype have a significantly poorer prognosis than those with Luminal/Her2+ or Luminal tumors (21). Patients with TN tumors have worse overall and disease-free survival rates (22), especially in patients with lung metastases. Even patients with non-metastatic TN breast cancer have a high early risk of developing BrM as a first site of recurrence (23), and worse survival after brain radiotherapy (24) than those with the non-TN phenotype. In these patients, BrM represents a significant adverse prognostic factor not only to overall survival but also to neurologic and radiosurgical survival (25).
We recently reported BrM biomarkers that discriminate breast carcinomas according to their likelihood of BrM progression, regardless of whether or not they expressed Her2 (26,27). Of these, GRP94 (94 kDa glucose-regulated protein), a signaling regulator and a major endoplasmic reticulum chaperone and FN14 (fibroblast growth factor-inducible protein) implicated in many intracellular signaling pathways, both have been implicated in the promotion of tumor proliferation and metastasis.
GRP94 has calcium binding properties that are conferring its major function in protein folding, assembly and degradation (28,29). Tumor hypoxia activates endoplasmic reticulum stress upregulating the unfolded protein response (30). The expression of GRP94 correlates with advanced stage and poor survival in many cancers (31,32).
In addition, FN14 is implicated in several signaling pathways that control the cancer hallmarks (33). Typically, reactive astrocytes produce proinflammatory cytokines, among them TWEAK (TNF-like weak inducer of apoptosis), a type II membrane protein which activates FN14 (34). The binding of TWEAK to FN14 is involved in regulating perivascular astrocytes and the blood-brain barrier interface (35). Moreover, FN14 has been involved in cachexia and the treatment with anti-FN14 antibodies improves body and muscle mass and adipose tissue in mice, increasing survival and general welfare (36).
Given these results, we hypothesized that the expression of FN14 and GRP94 could be used for early identification of the risk of breast cancer brain metastasis, whatever the molecular subtype. Thus, we studied their expression in breast cancer primary tumors according to their molecular subtype defined by Her2, ER, PR, and Ki-67 expression. Our results indicate that FN14 is the most useful predictive/prognostic biomarker of BrM in breast cancer patients with Luminal (Luminal A, Luminal B, and Luminal/Her2+) carcinomas. Moreover, in combination with GRP94, FN14 predicts also BrM progression in non-Luminal tumors.

Patients
We obtained 211 samples from patients diagnosed between 1989 and 2009 (Table S1 in Supplementary Material) at the
The overexpression of GRP94 and FN14 was categorized as positive when strong expression was detected and negative when no or weak expression was detected, to avoid false positives ( Figure 1). Morphologic diagnosis was performed with classical hematoxylin-eosin staining.
statistics Frequencies of categorical variables were compared among groups using the χ 2 -test or Fisher's exact test where appropriate. Brain metastasis-free survival (BrMFS) was estimated for each group using the Kaplan-Meier method and was compared among them using the Cox proportional hazards model, estimating their hazard ratio and 95% CI.
To evaluate the correlation between BrM and protein expression, immunostained samples were graded on a three-category scale as follows: negative, weak positive, and strong positive. The marker was classified as overexpressed only in strong positive samples to avoid false positives. Biomarker sensitivity and specificity, both singly and in combination, was assessed in both Luminal and non-Luminal patients. The biomarkers combinations were considered positive when at least one of them was positive and negative when all of them were negative.
A multivariate logistic regression analysis adjusted for covariates was carried out in both Luminal and non-Luminal groups to study in patients with BrM vs. NBrM the presence of biomarkers (GRP94 and FN14) in their primary tumor. The covariates used were as follows: age (≥50, 40-49, and <40), positive axillary nodes (0, 1-3, and ≥4), Her2 status (negative and positive), and presence of lung metastasis (no and yes), where the first category mentioned for each variable was the reference. We calculated the OR associated with the biomarker, its 95% CI and p-value.
In this analysis, the variable "triple negative (no, yes)" was not included as a covariate because all patients belonging to the Luminal group were "no" for this variable. Moreover, Her2 status was not included as a covariate when the combination (GRP94 + FN14 + Her2) was used as the biomarker. Values were considered significant when p was less than 0.05. Software used: R Core Team (37).

resUlTs clinical characteristics of Breast carcinoma subtypes and BrM involvement
We studied patient characteristics according to three different groups of progression patterns (Table S1 in Supplementary Material): brain metastases (BrM), with or without metastases (WoM) at other sites; non-brain distant metastases (NBrM), patients with metastasis in bones and/or liver and/or lungs and/ or non-regional lymph nodes, but not in brain; and patients WoM. The distribution of breast cancer molecular subtypes changed across the three groups of patients. The main distinctive parameters that characterized the BrM group were age, whereby below 50 years old was significantly different (p = 0.001); hormone receptor negativity [both ER (p < 0.0001) and PR (p < 0.0001)], an attribute of tumors that developed BrM in contrast to tumors from NBrM and WoM patients; and Her2 positivity and a high Ki-67 index (p = 0.01 and p < 0.0001, respectively). Other parameters, such as tumor size (p = 0.001), HG (p < 0.0001), and lymph node involvement (p < 0.0001) were similar among BrM and NBrM patients, but different in WoM patients.
More  Figures S1A,B in Supplementary Material). Moreover, the worst prognosis was found in patients with non-Luminal tumors (HR = 10.57, 95% CI: 5.60-19.96; and HR = 4.01, 95% CI: 2.52-6.38; respectively, p < 0.0001). These results indicate that the assessment of subtypes in our series provided an effective subclassification according to BrM progression in patients, similar to other reported series (11,14).  Table 2).

Fn14 and grP94 are Prognostic Biomarkers for BrM in luminal Tumors
First, we studied metastasis-free survival in our series according to molecular subtype ( Figure S2 in Supplementary Material), using the Kaplan-Meier method and the Cox proportional hazards model. The TN subtype was used as a reference group for comparative purposes. We analyzed the distribution of subtypes in NBrM patients ( Figure S2A in Supplementary Material). Differences between the TN and Her2-enriched subtypes were not statistically significant (HR = 0.81; 95% CI: 0.41-1.60; p = 0.53). By contrast, Luminal/Her2+ (HR = 0.41; 95% CI: 0.18-0.96; p = 0.039), Luminal B (HR = 0.33; 95% CI: 0.17-0.64; p = 0.0012), and Luminal A (HR = 0.16; 95% CI: 0.08-0.30; p < 0.0001) showed a significantly lower incidence of NBrM than TN breast carcinomas. Furthermore, in BrM patients, we found significant differences in BrMFS with regard to molecular subtype, using the TN subtype as the reference group ( Figure S2B  Next, we studied BrMFS in the Luminal group according to whether the tumor expressed FN14 or not (Figure 2A), and we found that overexpression of FN14 was associated with a reduction in BrMFS (HR = 8.25; 95% CI: 2.77-24.61; p = 0.00015). Although, different stratification of the non-Luminal group ( Figure 2B) according to FN14 expression was not statistically significant (HR = 1.74; 95% CI: 0.87-3.47; p = 0.11).

Fn14 is a Predictive Biomarker of BrM in luminal Tumors
Since overexpression of the Her2 gene is associated with a higher risk to develop BrM, we analyzed the sensitivity and specificity of FN14 and GRP94 expression to predict BrM and compared these parameters to the prediction given by Her2 in these patients ( Table 3). In patients belonging to the Luminal group (low risk of BrM a priori), FN14 and Her2 showed more specificity (89.60 and 89.84%, respectively) than GRP94 (55.56%) to predict BrM progression. However, GRP94 expression showed more sensitivity (76.92%) than FN14 and Her2 (53.85 and 46.15%, respectively) to predict BrM involvement.
On the other hand, in patients with a higher risk of BrM, such as non-Luminal ones, FN14 (88.89%) was the most specific protein to discriminate tumors that developed BrM, followed by Her2 (67.86%) and GRP94 (60.71%). GRP94 sensitivity was again higher (57.89%) than that obtained with FN14 or Her2 (31.58 and 30.77%, respectively).
Expression of the biomarker combination FN14 + GRP94 improved BrM risk assessment in Luminal patients compared with non-Luminal ones (sensitivity: 84.62 and 68.42%, specificity: 50 and 57.14%, respectively). The addition of Her2 yielded 84.62 and 71.79% sensitivity, and 45.24 and 42.86% specificity in Luminal with regard to non-Luminal, respectively ( Table 3). The   In each case, these combinations were considered positive when at least one of the assessed molecules was positive and negative when all of them were negative. b Differences in number or patients (N) are due to those patients whose biomarker assessment (singly or in combination) was unknown. The high specificity shown by FN14 and Her2 when evaluated singly was lost in combination, suggesting that both biomarkers stratified different patient subgroups of BrM risk.
To assess the usefulness of FN14 and GRP94 as independent risk factors of BrM, a multivariate analysis was performed, including age, axillary node involvement, Her2 status, and presence of lung metastasis as covariates in the analysis ( Table 4).  In non-Luminal patients, only the combination of FN14 and GRP94 positivity was significantly predictive of BrM progression (95% CI: 1.19-13.65, p = 0.025), with a 4.04-fold likelihood to develop BrM ( Table 4). These results are in consonance with our previous results reporting the usefulness of both biomarkers to predict BrM in TN breast cancer patients (27).

Outcomes showed that patients belonging to the
In summary, our study reveals that (1)  Since stratification impairs statistical consistency, further multicentre studies with wider amount of patients are needed to reinforce the results.

DiscUssiOn
The routine analysis of ER, PR, Ki-67, and Her2 status in breast tumors can predict relapse, providing the standard approach for clinical decision-making in the adjuvant setting (17)(18)(19). However, these procedures are insufficient to predict BrM.
This study provides evidence that a subset of breast cancer patients with a better prognosis a priori, such as patients with Luminal carcinomas, with or without Her2 positivity, can be stratified by their likelihood to develop BrM if FN14 is overexpressed (OR = 36.70). Thus, the clinical use of FN14 expression might facilitate a preventive strategy for patients at high risk for BrM progression and will improve the design of trials aimed at its prevention. Moreover, the combined assessment of both FN14 and GRP94 proteins shows a higher benefit in risk evaluation of BrM progression, especially in non-Luminal patients (4.04-fold), independently of Her2 status (OR = 2.45). Therefore, the use of FN14 and GRP94 expression at early diagnosis might stratify those BrM patients prone to BrM.
Many studies have reported risk factors for BrM, including Her2 positivity, ER negativity, high proliferative activity, young age and lymph node involvement (38,39). FN14 and GRP94 comprise a diagnostic tool capable of predicting BrM independently of these classical clinical and pathological parameters. The long-term BrM-free survival of Luminal group patients when biomarkers are negative suggests the usefulness of including both biomarkers to stratify patients that might benefit from magnetic resonance imaging (MRI-Gd) follow-up. This should be considered for at least 7 years (about 80 months) after diagnosis, the period after which BrM relapse in the Luminal group stabilizes (Figure 2). Therefore, we suggest that intrinsic subtypes of breast cancer plus FN14 and GRP94 expression can provide a reliable assessment of BrM risk, facilitating early diagnosis through follow-up of the patient's evolution (Figure 3). Even patients from the non-Luminal group could benefit from stratification using FN14 and GRP94 biomarkers. This is not surprising because in combination, they are good predictors of BrM progression in TN breast carcinomas (27). Consequently, if these findings were confirmed in further studies, it would also enable us to apply a specific clinical and therapeutical algorithm to improve breast cancer patients' follow-up.
Although the armamentarium available for BrM treatment is limited, there are reasons to be optimistic because emerging therapies have shown promise in preclinical and early clinical settings (40)(41)(42). Moreover, a protocol that included an MRI procedure might provide indications for early surgery and/or radiosurgery when BrM is small to minimal and/or for the design of a new approach in prophylactic systemic protocols (e.g., to replace or add another drug and/or biological compound that crosses the blood-brain barrier to avoid the growth of a clinical BrM) as well as for the design of a new protocol as a prophylactic approach.
Patients with Her2 positivity in primary tumor are usually treated with trastuzumab after delivering chemotherapy, obtaining a better systemic response (43). In our series, 99% of patients did not receive trastuzumab as adjuvant therapy. Thus, an interesting approach would be to study the relationship between FN14 and GRP94 expression and BMFS in those patients belonging to the Lum/Her2+ and Her2-enriched subtypes who have received trastuzumab. In our multicenter series, GRP94 and FN14 expression might improve breast cancer survival by predicting BrM. In particular, FN14 has a similar sensitivity and specificity to that of Her2.
Immunophenotypic changes associated with antitumor activity have been observed with anti-TWEAK antibody treatment in mice and a phase I multicenter trial of RG7212 monotherapy in patients with FN14-expressing advanced solid tumors has been initiated, with good tolerability and favorable pharmacokinetics (44). Therefore, these molecules might be good candidates to develop new drugs to treat or prevent BrM according to the tumor-associated risk of breast cancer patients. In itself, BrM is an exclusion criterion for most prospective trials, limiting the possibility of developing new therapies (45). Moreover, therapies are usually started when symptoms appear, limiting treatment options and success (41). We propose a new classification that provides a standard approach for clinical decision-making about CNS metastases at early diagnosis when adjuvant chemotherapy and radiosurgery are more effective (46). Furthermore, evaluation of new and more specific biomarkers in primary tumor may be a promising field of research due to the high impact that these might have in the future as regards facilitating the design of new therapeutic strategies to either prevent or treat this life-threatening event.

eThics sTaTeMenT
This study was approved by Comité etico de Investigación Clínica del Hospital Clínico de Barcelona.

aUThOr cOnTriBUTiOns
AM-A designed the study hypothesis according to both ER status and biomarkers expression, collecting all available clinical and pathological information from patients' reports and designing the several clinicopathological variables in a new wider and more complete database. VH participated in the follow-up of tissue microarrays and collecting and organizing the whole information emerged from biomarkers expression. FM and RB, as radiation oncologists, and AU, MG-G, and MS, as medical oncologists, contributed with a wide registered clinical follow-up of patients and they offered important clinical data from clinical reports. NB, NV, XA, and EC, as pathologists, contributed with the tissue microarrays analysis and categorizing the level of expression of biomarkers in all studied patients. DC carried out all statistic analysis. AS is the leader and responsible of the project, leading, coordinating,