JCM, Vol. 11, Pages 7151: Preoperative Risk Stratification of Increased MIB-1 Labeling Index in Pituitary Adenoma: A Newly Proposed Prognostic Scoring System

1. IntroductionPituitary adenomas (PAs) are considered to be predominantly benign neoplastic diseases and account for 15% of all central nervous system (CNS) tumors [1]. The prediction of the natural course of PAs is very challenging because of a vast heterogeneity of influencing factors such as clinical signs, neuropathological characteristics, proliferative activity, and the growth pattern with potential invasion into surrounding anatomical structures [2]. The reduction of mass effect to relieve clinical signs (e.g., loss of vision) and the prevention of such dysfunctions, in case of further tumor growth, are the primary indications for surgical resection. Due to the low-risk profile of the transsphenoidal approach, neurosurgical resection is the treatment of choice in all cases except for prolactinoma. Elevated cellular proliferation is the main avenue of oncogenesis [3]. The Molecular Immunology Borstel (MIB-1)-1 labeling index is an established immunohistochemical diagnostic to identify nuclear structures that are found in cells during proliferation. The Ki-67 antigen can be detected in the nuclei of neoplastic cells in the following phases of mitosis and cell division cycle: G1, S, and G2. Hence, the labeling of this antigen is a simple method to identify the growing fraction of tumor tissue [4,5,6]. Moreover, the MIB-1 labeling index was suggested to distinguish PAs and pituitary carcinomas [7]. Elevated MIB-1 labeling indices were reported to be associated with a higher probability of PA progression [8]. However, the routine examination and inclusion of the MIB-1 labeling index for the classification of PAs are still highly debated. A systematic review analyzing 28 studies on the MIB-1 labeling index, identified 18 studies that found increased MIB-1 labeling indices in recurrent PAs, whereas 10 studies found no correlation [9]. To achieve an optimum long-term outcome regarding PA progression, reliable preoperative evaluation, adequate medical information about the goals of surgical therapy, and surgical treatment with preservation of function are of paramount importance. Nevertheless, the MIB-1 labeling index is not available in the preoperative phase of the oncological decision-making process and interactive doctor-patient consultation. An increased MIB-1 labeling index implies an increased need to achieve maximum cytoreductive surgical therapy. To date, there are approaches such as machine learning models using T2 weighted magnetic resonance imaging (MRI) or nomograms combining clinical, demographic, and imaging characteristics to preoperatively estimate the MIB-1 labeling index [10,11]. However, no simple and quick-to-use scoring sheet exists so far to enable an estimation of the MIB-1 index prior to surgical resection. Hence, we have investigated our patient population of sporadic PAs regarding preoperative clinical signs, laboratory inflammatory and hormone markers, and imaging risk factors for an elevated MIB-1 labeling index. Moreover, the present investigation was intended to devise a proposal for a new scoring sheet to display demographic data, laboratory inflammatory data, and endocrine functioning to identify PA patients who are at risk of having an elevated MIB-1 labeling index. 4. DiscussionResidual tumor volume, invasive growth, and young age are known poor predictors regarding regrowth in sporadic PAs [21]. An increased MIB-1 labeling index is considered a molecular marker for aggressive tumor biology, which implies an increased risk for invasive growth and shortened time to tumor progression in PAs [22]. Several investigations have revealed that an elevated MIB-1 labeling index is associated with increased proliferative activity and aggressive nature in PAs [23]. In a typically elective setting regarding the hospital admission of PA patients, it is essential that PA patients as well as their relatives are provided with a maximally profound physician–patient dialog. Nevertheless, the MIB-1 labeling index reflecting the proliferative activity cannot be used in the preoperative therapy planning regarding the extent of resection or weighing-up of surgical and conservative treatment. The present investigation suggests a score to estimate an elevated MIB-1 labeling index. This system strives to identify patients at risk of a high MIB-1 labeling index and uses four preoperative characteristics. Moreover, this scoring template might enable the identification of high-risk patients regarding a shortened time to PA progression.

Our results can be described as follows: (1) a threshold MIB-1 labeling index value of 4% enables a risk stratification regarding recurring and nonrecurring PAs in those who underwent complete resection; (2) increased IGF-1, young age at diagnosis, increased ACTH, and increased plasma fibrinogen were significantly associated with an elevated (≥4%) MIB-1 labeling index; (3) at least one characteristic among young age, increased plasma fibrinogen, and increased ACTH combined with an increased IGF-1 seems to identify patients at risk of an elevated MIB-1 labeling index; and (4) the presence of at least two variables among young age, increased plasma fibrinogen and increased ACTH in combination with an increased IGF-1 results in an increased risk for shortened time to PA regrowth in completely resected PAs.

In the present investigation, we created a Kaplan–Meier chart for the probability of PFS in completely resected PAs stratified MIB-1 index. The optimum threshold was set at ≥4% based on the literature [20]. We found that an MIB-1 index ≥4% is significantly associated with a shortened time to PA progression in completely resected PAs. Preoperatively, an accurate estimation of the MIB-1 labeling index might guide neurosurgeons, endocrinologists, and radiotherapists to provide a tailored treatment schedule. The MIB-1 labeling index was also integrated into a five-tiered classification system regarding invasive growth and proliferative potential. This mentioned classification system for pituitary adenomas underwent an external validation in four independent cohorts [24].The insulin-like growth factor-1 is synthesized in the liver and acts as a mediator of GH. Elevated serum concentrations of IGF-1 are broadly accepted as the screening test of choice in the diagnostic workflow of patients with acromegaly [25]. Several studies suggested the MIB-1 labeling index as an important clinical outcome parameter regarding tumor control in somatotroph PAs. However, previous studies found no significant association between the MIB-1 labeling index and IGF-1 serum levels [26]. Epidemiological studies suggested a strong association between circulating serum IGF-1 levels and the risk of several cancers such as breast cancer. Furthermore, IGF-1 signaling is potentially associated with cancer progression [27,28]. In breast cancer, IGF-1 levels were found to strongly correlate with the MIB-1 index [29]. Several potential pathophysiological mechanisms might be responsible for our findings. IGF-1 has multiple functions and different roles in the development and progression of several diseases. In a physiological setting, IGF-1 inhibits apoptotic effects and supports cell survival, whereas, in pathophysiological conditions, IGF-1 can enhance cancer progression or increase the number of adipocytes [30,31,32,33,34]. However, our identified optimum cutoff value of IGF-1 regarding the identification of an elevated MIB-1 index is within the physiological reference interval. Furthermore, the clinical symptom of acromegaly itself was no predictor of an increased MIB-1 labeling index. Therefore, the identified association seems to be paradoxical. Nevertheless, those findings suggest that the created scoring system is not exclusively designed for a cohort of somatotroph PAs, and it might be generally transferrable to all primary PAs. Further investigations with a special focus on the association between IGF-1 and PA progression are necessary.Furthermore, age was also found to have an inverse association with an increased MIB-1 labeling index. This result is also in line with the findings of Cai et al. [11]. They found that the mean age of patients with an MIB-1 labeling index ≥3% is significantly higher compared to those with an MIB-1 labeling index 35,36]. Furthermore, advanced age was found to effectively inhibit tumor recurrence in nonfunctioning PAs according to the results of the multivariable Cox regression analysis in a retrospective study of 145 patients [37]. We found an association between ACTH levels and increased MIB-1 index. This finding is also supported by the study of Pizzaro et al. [38] which measured the MIB-1 labeling index in 159 PAs and revealed that ACTH-secreting adenomas have significantly higher MIB-1 labeling indices [38]. Moreover, Mastronardi et al. [39] found that ACTH-secreting PAs have a mean MIB-1 labeling index of 5.88 +/− 9.13%, whereas other hormone-secreting or nonfunctioning PAs had a mean MIB-1 labeling index of 2.33 +/− 2.4%.Plasma fibrinogen was found to be independently associated with the MIB-1 index. Plasma fibrinogen is linked to the interleukin-6 (IL-6) gene promoter and is induced by the autocrine functioning of pituitary adenoma cells [40]. In a pathological study evaluating the localization and expression of IL-6, IL-6 receptor, and the signal-transducing subunit (gp130) using immunohistochemistry and reverse transcription PCR, IL-6 was predominantly expressed in ACTH- and FSH/LH-secreting cells. Furthermore, IL-6 might function in GH and prolactin-secreting cells through paracrine and endocrine pathways, whereas IL-6 may function in FSH-secreting PA cells in an autocrine manner [40]. The analysis of the cytokine secretome in 24 PAs in primary cultures using an immunoassay panel with 42 cytokines found that PAs with a deleterious immune phenotype including dense macrophage infiltrates and a cluster of differentiation (CD) 4+ T lymphocytes had higher MIB-1 labeling indices. Hence, it was suggested that PA-derived chemokines might enhance the recruitment of macrophages, neutrophils, and T cells into the tumor tissue resulting in a more aggressive behavior [41]. Hence, the systemic inflammatory burden was also identified to be higher in PA patients compared to healthy individuals [42]. The secretome of PAs also includes hormones such as ACTH which are secreted into the circulation and significantly modulate the hematopoiesis as well as circulating immune cells contributing to the degree of systemic inflammatory burden. This pathophysiological condition of PA’s secretome-induced inflammation is well-known for Cushing´s disease [43,44]. Marques et al. [45] performed a retrospective evaluation of 424 PA patients and investigated the usefulness of blood-based inflammation markers to predict the disease course. They found that a score using serum inflammation markers might predict invasive and refractory PAs. Nevertheless, it has to be reminded that there is also a potential confounding effect because systemic inflammatory markers might be influenced by further comorbidities and corticosteroid treatment. Furthermore, plurihormonal PAs were not associated with an increased MIB-1 index in our cohort. However, it also has to be reminded that the determination methods of plurihormonal PA using clinical signs, serum hormone concentrations, and pathological results are increasingly discussed [46]. In the present series, the plurihormonal PAs are determined using immunohistochemical methods as recommended by other investigators [46,47]. Nevertheless, the immunohistochemical detection of hormones might not always result in a laboratory increase in serum hormone concentrations or clinical endocrine signs. Hence, it is debatable whether this classification has some clinical implications. This phenomenon might be explained by the fact that hormones secreted by the PA are biologically inactive or that they lost their functioning after entry into the blood system [46,48]. Moreover, in the present series, we did not identify an association between p53 expression and an increased MIB-1 labeling index. The prognostic value of determining p53 expression in PAs is controversial and there is no recommendation to routinely include it in the classification workflow according to the WHO [16]. Furthermore, different staining methods and heterogeneous cohorts (proportions of micro- and macroadenomas) also resulted in a broad range from 17% to 60% regarding the frequency of observed p53 expression [49,50].

The novel FATE score in the present investigation provides a score to preoperatively estimate an increased MIB-1 labeling index and might guide physicians in estimating the risk of tumor progression in completely resected PAs. This risk index may facilitate the preoperative treatment planning and the patient–physician dialog because neuropathological characteristics are only available after surgery so far. PA patients with an elevated FATE score (≥3) who prefer a conservative regimen of an incidental PA have to be informed about a more stringent follow-up imaging schedule regarding the time intervals. Furthermore, the FATE score was also found to be significantly associated with the probability of PFS in completely resected PA patients. Therefore, the identification of the FATE score as a potential sufficient surrogate marker for the MIB-1 index might facilitate devising the treatment strategy for PA patients and facilitate tailored postoperative follow-up scheduling.

LimitationsSeveral limitations are present in this investigation. Despite the data being acquired from a selective and homogeneous population, the retrospective nature of this investigation suffered from a monocentric experience. Furthermore, other inflammatory markers such as cytokines, which may provide more profound details regarding the interaction between inflammation and proliferation, were not available in this retrospective investigation. Moreover, there are potential interlaboratory differences regarding the determination methods of pituitary hormone concentrations and also the determination methods of the MIB-1 labeling index in cancer tissue (e.g., digital imaging analysis, hotspot, average method) [51]. Those limitations must be considered before our results can be transferred to clinical practice or external validation. Therefore, a multicentric prospective trial with a thorough, homogeneously balanced study protocol should provide external validation for this score to enable its reliable integration into the healthcare of PA patients.

留言 (0)

沒有登入
gif