Common Prognostic Scoring Systems for Patients Presenting with Brain Metastases

A B S T R A C T

Brain metastases (BM) of various primaries merely remain the most prevalent type of intracranial tumors, and approximately 25% of all cancer patients are diagnosed with this poor prognostic disease condition somewhere during their treatment course. Contingent upon the general wellbeing status of the potential patient, currently available major treatment options typically include palliative radiotherapy, chemotherapy, and best supportive care. Various published studies have convincingly shown the likelihood to stratify BM patients into particular prognostic gatherings according to the conceivable combinations of multiple patients- and tumor-related characteristics; namely the prognostic scoring systems, which might be useful in the accurate prediction of survival, and thusly, the appropriate choice of the best-fit treatment alternative. In this present article, we meant to review the pros and cons of the as of now accessible and broadly acknowledged prognostic scoring systems for BMs and their clinical values.

Keywords

Brain metastasis, scoring systems, prognosis, survival

Introduction

Approximately 25% of all cancer patients are diagnosed with brain metastases (BM), which tragically increases up to 64% throughout their treatment course [1-4]. The exact incidence of the newly diagnosed BM is unknown, but it is estimated to be 3 to 10 times the incidence of newly diagnosed primary brain tumors [5, 6]. The BM incidence assuredly appears to further rise in the foreseeable future as a tangible result of longer survival expectations following the successful implementation of more sophisticated diagnostic imaging modalities and earlier commencement of effective local/regional and systemic anticancer interventions.

Hypothetically, all aggressive cancers may metastasize to the brain, however, the majority of BMs stem from the lung cancers (36-64%), breast cancers (15-25%), malignant melanoma (15-25%), and gastrointestinal cancers (5-10%), with an unknown primary in further 10-15%, respectively [7-9]. Malignant melanomas have the highest penchant for BM amongst all primary malignant tumors [10]. The distribution patterns of BMs usually follow the natural brain bloodstream pathways: 80%, 15%, and 5% in the cerebral hemispheres, cerebellum, and brainstem, individually [11]. Moreover, most BMs typically emerge at the intersection zones between the gray and white matters of the brain, presumably as the desired result of the natural localization of the capillary beds at these regions [12]. Currently, essentially 50% of all BMs are multiple at diagnosis possibly owing to the frequent utilization of highly sensitive and specific magnetic resonance imaging (MRI).

Presentation with BM indicates an adverse prognostic condition with expected survival duration of usually less than a year, yet the prognosis of BM patients may differ broadly due to multiple factors; including the age, performance status, total number and volume of the metastatic lesion(s), treatment modality utilized against the BM, extracranial disease status, and histology of the primary malignancy. Various tumor-specific molecular factors, pathologic biomarkers, and driver mutations have also demonstrated significant prognostic utility in BM of lung-, breast-, hepatocellular carcinomas, and malignant melanomas [13-17]. In past investigations, numerous researchers have blended these identified factors in various manners and created prognostic scoring systems to accurately anticipate the survival of BM patients and to properly guide their treatments in an ideal way. Considering the fundamental significance of the stratification of BM patients according to the widely accepted prognostic factors which may guide for the selection of the best-fit treatment modality and its intensity, the present article aimed to succinctly summarize the pros and cons of accessible prognostic systems developed for BM patients.

Prognostic Factors

Brain metastases diagnosed at any phase of cancer treatment or follow-up is typically perceived as an incredibly poor prognostic factor for almost all cancers with an expected median survival of 2 to 7 months from the diagnosis [18]. Most patients die because of widespread systemic disease rather than the BM, but still, a particular group may survive for longer durations or even may cure if appropriately managed. For the treatment of BM, such patients are potentially suitable candidates for more aggressive and potentially less neurotoxic treatment strategies like neurosurgery or stereotactic radiosurgery (SRS) rather than the more toxic long-course whole-brain radiotherapy (WBRT) [4]. In this patients’ group, comprehensive assessment of independent prognostic factors and their unique blends may undoubtedly guide the proper selection of best-fit treatment strategies which may result in the excellent preservation of neurologic functions with likely protraction of survival times.

Besides the individual usage of the previously mentioned clinical variables, a comprehensive combination of prognostic factors for BM patients may thoroughly stratify them into specific groups with significantly distinct brain control and survival outcomes. To date, many demographic and clinical variables have been extensively examined for their predictive and prognostic roles in BM patients (Table 1).

Table 1: Prognostic factors for patients presenting with brain metastases.

Factor	Tumor site
Age	Common
Gender	Common
Performance status	Common
Location of BM	Common
Number of BM	Common
Size of BM	Common
BM velocity	Common
Volume of BM	Common
Neurologic deficit status	Common
Extracranial disease status	Common
Histology of BM	Radioresistant vs. radiosensitive
ER/PR status	Breast
HER-2 status	Breast
EGFR status	Non-small cell lung
EML4-ALK status	Non-small cell lung
BRAF status	Malignant melanoma
Caveolin-1	Non-small cell lung
Peritumoral edema status	Common
Radiologic features	Common
Interval from primary diagnosis	Common
Radiotherapy technique (SRS vs. others)	Common
Type of systemic therapy	Non-small cell lung, breast, malignant melanoma

BM: Brain metastasis; ER: Estrogen receptor; PR: Progesteron receptor; HER-2: Human epidermal growth factor receptor 2; EML4-ALK: Echinoderm microtubule associated protein-like 4 and anaplastic lymphoma kinase; BRAF: v-Raf murine sarcoma viral oncogene homolog B.

Among them, the performance status, age at presentation, neurologic function status, tumor histology, the primary tumor control status, the presence/absence of extracranial metastases (ECM), and the number and size of BM were distinguished to be clinically meaningful [4, 19-21]. Moreover, independent investigators proposed various prognostic scoring frameworks for patients with BM by utilizing different combinations of these prime factors with variable approaches (Table 2), as discussed below:

I Recursive Partitioning Analysis Scoring System

In 1997, Gaspar et al. proposed the first prognostic scoring system for BM patients by utilizing the recursive partitioning analysis (RPA) methodology in 1,200 patients enrolled on three RTOG studies and treated with WBRT [22-25]. This scoring system discovered the KPS, age, primary tumor control status, and the status of ECM as the significant correlates of survival among the 21 influential factors analyzed. Accordingly, the RPA class I patients had the best prognosis with a median OS of 7.1 months, while the RPA II and RPA III patients demonstrated stepwise OS decrements with respective 4.2- and 2.3-months duration. Nonetheless, the RPA classification has some critical drawbacks including the ignorance of patients with KPS≤ 60 by fixing its utility for patients presenting with pre-WBRT KPS of≥70, significantly large variations between WBRT doses of the trials and prohibition of the number of BM from RPA analysis. Further constraining its broad and convenient routine usage, this system amasses most patients mainly in the RPA class II, as the dominant class. One further basic impediment of the RPA classification system is the fact that it incorporates all KPS <70 patients into a single class, namely the RPA class III regardless of the accompanying clinicopathological factors. However, these additional factors may alter the survival times hugely either negatively or positively. Landing support on these adverse remarks, Nieder et al. in a later study incorporating 113 BM patients underlined that there were no meaningful survival differences between the patients in classes II and III (3.6 versus 4.2 months; P>0.05) after a total dose of 30 Gy WBRT in 3 Gy daily fraction doses [26].

Table 2: Comparison of frequently utilized prognostic scoring systems for brain metastases.

	RPA (n=1200)	SIR (n=65)	Rotterdam (n=1292)	BSBM (n=110)	Rades (n=1797)	GGS (n=479)	GPA (n=1960)	DS-GPA (n=4259)
Age	Yes	Yes			Yes	Yes	Yes	Yes
Performance status	KPS	KPS	ECOG	KPS	KPS	KPS	KPS	KPS
Primary control	Yes	Yes		Yes
ECM	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Number of BM		Yes			Yes	Yes	Yes	Yes
Volume of BM		Yes
Steroid response			Yes
Tumor histology						Yes	Yes	Yes
Interval from diagnosis to RT					Yes
ER/PR status								Yes
HER-2 status								Yes
EGFR status								Yes
EML4-ALK status								Yes
BRAF status								Yes
AFP								Yes
Child-Pugh-Score								Yes

Note: Empty spaces represents for ‘No’.

RPA: Recursive Partitioning Analysis Scoring System; SIR: Score index for radiosurgery; BSBM: Basic score for brain metastases; GGS: Golden grading system; GPA: Graded prognostic assessment; DS-GPA: Disease specific graded prognostic assessment; ECM: Extracellular matrix; BM: Brain metastasis; RT: Radiotherapy; ER: Estrogen receptor; PR: Progesteron receptor; HER-2: Human epidermal growth factor receptor 2; EGFR: Epidermal growth factor receptor; EML4-ALK: Echinoderm microtubule associated protein-like 4 and anaplastic lymphoma kinase; BRAF: v-Raf murine sarcoma viral oncogene homolog B; AFP: Alpha fetoprotein.

II Score Index for Radiosurgery

The score index for radiosurgery (SIR) system created by Weltman et al. involved the number of BM, the volume of the largest BM, the location of BM, and post-SRS WBRT in addition to the essential components of RPA [27]. The authors proposed the SIR as a comparably more reliable post-SRS survival predictor than the RPA system. Even though the SIR was later approved by further research which incorporated patients treated with neurosurgery alone or with additional WBRT, yet couldn't gain a wide arena for its daily usage in the oncology communities mainly due to the small population size of the seminal study (N=65) and probably for a more considerable extent due to the need for clinically impractical and time-consuming comprehensive workup for evaluation of the systemic disease in this classification method.

III Rotterdam Score

Lagerwaard et al. identified prognostic factors in 1292 patients with BM to accurately determine proper subgroups of patients suitable for wise selection in future trials [28]. Besides the well-recognized KPS, age, control of primary tumor, and the status of ECM the authors further demonstrated that the response to steroids, serum lactate dehydrogenase, sex in lung primary, number of brain metastases, and site of the primary tumor comprised the factors to be associated with outcomes. The most powerful three factors, namely the performance status, response to steroids, and evidence of ECM were used for the simpler lamination of BM patients. However, because most cancer centers do not retain adequate records for the response to steroids, the Rotterdam scoring system could never attain a broad application arena in routine oncology practice.

IV Basic Score for Brain Metastases

The basic score for BM (BSBM) was generated as a novel classification by Lorenzoni et al. in 2004 and incorporated only three factors to simplify the scoring system: KPS, primary tumor control status, and the presence of ECM [29]. Analysis of 110 BM patients undergoing SRS showed excellent agreement between the BSBM and SIR for accurate prognostic stratification of the study population. Lorenzoni’s published findings were later confirmed by further evaluation of BSBM in patients receiving WBRT plus neurosurgery and WBRT with/without SRS, However, alike the SIR system, the BSBM framework was also handicapped by the initial work’s limited cohort size (N=110), which may render the assessment of more compact groups formidable because of the large confidence intervals [13, 26, 30].

V Graded Prognostic Assessment

Outcomes of the randomized RTOG 95-08 trial exhibited that the number of BM was a significant prognostic factor in patients with 1 to 3 BM who received WBRT alone or WBRT plus SRS boost [31]. However, the number of BM was not included in either of the published RPA, BSBM, and Rotterdam scoring systems [21]. On this account, the graded prognostic assessment (GPA) which incorporated age, KPS, ECM, and the number of BM in the scoring framework by analyzing the outcomes of 1960 patients treated with WBRT alone, WBRT plus radiosensitizers, or WBRT plus SRS in five RTOG trials (RTOG 7916, 8528, 8905, 9104, and 9508) was proposed as a novel prognostic scoring system in 2007 [32]. This novel system scored each factor as 0, 0.5, or 1.0 and stratified patients into four prognostic groups according to the resultant sum score of all 4 factors. Patients with the best prognosis were signed to GPA 4. The median OS was 2.6, 3.8, 6.9, and 11 months in GPA 0-1, GPA 1.5-2.5, GPA 3.0, and GPA 3.5-4 score groups, respectively. Considering these outcomes, the creators of GPA inferred that the GPA was the least subjective, most quantitative, and easiest to use scoring system compared to the preceding RPA, SIR, and BSBM systems. Following its publication, the GPA scoring system became a commonly preferred tool for prognostic stratification of BM patients as further studies confirmed the validity of the GPA system shortly after its announcement [33-35].

VI Disease-Specific Graded Prognostic Assessment

In 2008, Golden et al. analyzed the outcomes of 479 newly diagnosed BM patients of various primaries treated with SRS and demonstrated that the primary tumor type provided significant prognostic impact on the survival results [36]. Principally based on this challenging finding Sperduto et al. assessed the outcomes of 4,259 patients from 11 institutions to define disease-specific GPA (DS-GPA), and the authors exhibited that different variables had significantly distinct influences on the survival of patients in specific tumor types [37]. In 2012, Sperduto and colleagues in a large database of women presenting with BM of a breast primary refined their previous GPA scoring system for this particular patient’s group [16]. In this multi-institutional study, significant prognostic factors by multivariate Cox regression and RPA were determined to be the KPS, HER-2, ER/PR status, and the interaction between ER/PR and HER2. RPA showed age was significant only for patients with KPS 60 to 80. The median OS for GPA scores of 0-1.0, 1.5-2.0, 2.5-3.0, and 3.5-4.0 were 3.4, 7.7, 15.1, and 25.3 months, respectively (p<0.0001). Additionally, being ER (+)/PR (+) improved median OS from 6.4 to 9.7 months among HER-2(-) patients, while being ER (+)/PR (+) improved median OS from 17.9 to 20.7 months in HER-2(+) patients.

For BM of the malignant melanoma, the first and original Melanoma-GPA was based on data from 483 patients diagnosed between 1985 and 2005 [37]. The initial investigation identified the KPS and the number of BM as the unique factors with significant influence on the survival outcomes. Its recently published multi-institutional update involved 823 malignant melanoma patients with BM [15]. In this refined index, namely the ‘melanoma molecular-GPA (Melanoma mol-GPA)’, the molecular markers were also investigated for their impact on results. In multivariable analyses; age, KPS, ECM status, number of BM, and BRAF status were identified to comprise the five significant prognostic factors for survival. The median OS times for patients with Melanoma mol-GPA of 0-1.0, 1.5-2.0, 2.5-3.0, and 3.5-4.0 were 4.9, 8.3, 15.8, and 34.1 months (P<.0001 between each adjacent group), respectively.

The original DS-GPA for BMs from NSCLC was depended on four factors identified in 1833 patients and incorporated the age, KPS, ECM, and number of BMs [37]. In the more recent updated version, 2186 NSCLC patients (1521 adenocarcinoma and 665 non-adenocarcinoma) with a newly diagnosed BM were included and patients were furthermore examined for their molecular marker status: Lung molecular GPA (Lung molGPA) [14]. Noteworthy prognostic variables included the original four factors of the DS-GPA and added two new factors: EGFR and ALK alterations in patients with adenocarcinoma with no respect to the mutation status for non-adenocarcinoma cases. The median OS for the updated investigation accomplice was 12 months, and those with NSCLC-adenocarcinoma and Lung-molGPA scores of 3.5-4.0 had a median survival of almost 4 years.

For renal cell carcinoma (RCC) patients presenting with BM, the first DS-GPA distinguished the KPS and the number of BMs as the unique factors to significantly altering the OS outcomes in a group of 286 patients [37]. Recently, the same group identified additional prognostic factors in a larger cohort of 711 patients and updated the original Renal GPA [38]. In the revised Renal GPA; KPS, number of BM, ECM, and hemoglobin were discovered as the four most powerful variables. The median OS for Renal GPA groups 0-1.0, 1.5-2.0, 2.5-3.0, and 3.5-4.0 were 4, 12, 17, and 35 months (p<0.05, for each intergroup comparison), individually.

For gastrointestinal cancers, Sperduto et al. observed that the KPS was the key determinant of survival in the initial DS-GPA, with median OS times of 3.1, 4.4, 6.9, and 13.5 months for GPA groups 0-1.0, 2.0, 3.0, and 4.0, respectively [37]. Later the unique gastrointestinal system site specific GPA investigation was reported by Lim et al. in 2014 for BM of hepatocellular carcinomas [39]. In this study, the authors retrospectively reviewed the data from 118 hepatocellular carcinoma patients with newly diagnosed BM between 1985 and 2011, and created hepatocellular carcinoma GPA index by including the number of BM (single: 0.5, multiple: 0 points), alpha-fetoprotein (<400 ng/mL: 0.5, ≥400 ng/mL: 0 points), and Child-Pugh-Score (A: 3, B: 2, C: 0 points). The investigators could not demonstrate any values for age, sex, performance status, and time interval from initial diagnosis to development of BM, but reported that the median OS durations were significantly different when the hepatocellular carcinoma GPA was implemented: 1.7, 3.2, 7.9, and 27.0 weeks for hepatocellular carcinoma GPA scores of 0-1.0, 1.5-2.5, 3.0-.5, and 4.0, respectively (P<0.001).

VII Rades Prognostic Score

This comprehensive framework was generated by Rades et al. in 2011 [40]. In their investigation, 1,797 patients were randomly assigned to either of the test (n= 1,198) or the validation gatherings (n= 599). Two scoring frameworks were developed; one for intracranial control (IC) and another for OS. In multivariate analyses; age, performance status, ECM, the interval from the tumor diagnosis to RT, and the number of BM were found to be significantly connected with OS. Tumor type, performance status, interval, and number of BM were associated with IC. In the test group, 6-month IC rates were 17% for 14-18 points, 49% for 19-23 points, and 77% for 24-27 points (p<0.0001). IC rates in the validation group were 19%, 52%, and 77%, respectively (p<0.0001). In the test group, 6-month OS rates were 9% for 15-19 points, 41% for 20-25 points, and 78% for 26-30 points (p<0.0001). Corresponding OS rates in the validation group were 7%, 39%, and 79%, respectively (p<0.0001).

VIII Golden Grading System

The Golden Grading System (GGS) was developed by Golden et al. in 2008 [36]. In this system, the creators assessed the information acquired in 479 patients who experienced SRS with or without WBRT from 1991 to 2005 for newly diagnosed BM. Four groups were analyzed: 1) all locales consolidated, 2) breast, 3) lung, and 4) malignant melanoma primary sites. A multivariate examination of every essential site joined exhibited that the age <65 years, KPS ≥70, no ECM, and ≤3 BM were linked with longer OS, while primary tumor control was most certainly not. In the subgroup analysis of patients with breast, lung, or malignant melanoma primaries, favorable factors included only primary tumor control for breast-; age <65 years, no extracranial metastases, and ≤3 BM for lung-; and KPS ≥70, primary tumor control, and ≤3 BM for malignant melanoma primaries, respectively. The median OS for ≤3 versus > 3 BM was 15.6 and 16.9 months for breast, 16.5 and 11.3 months for lung, and 9.0 and 5.7 months for malignant melanoma gatherings.

Discussion

Survival of patients with BM has significantly prolonged with the valuable addition of novel targeted agents, immunotherapeutic, and locally ablative SRS to the conventional systemic chemotherapy and palliative RT. Since it is arduous for most agents to penetrate the blood-brain-barrier and achieve efficient concentrations in the cerebrospinal fluid the incidence of BM is ascending in parallel with the enhanced survival times. Enthusiastically supporting this critical observation, approximately 1/4 to 1/3 of all recurrences manifest in the form of the brain only relapses in radically treated stage III non-small cell lung cancer patients, the so-called oligometastatic state, if not all. As newer therapeutic agents are continuously added to the arsenal of oncologic treatment, it is pivotal to stratify BM patients into significantly distinct prognostic groups to handle them with the accessible best-fit option or spare some others from the futile toxicities of various aggressive treatment maneuvers.

The prognostic scoring systems for patients presenting with BM are useful tools in the accurate prediction of their survival outcomes and comforting assurance of the best treatment decision. Thusly, patients with the expected good prognoses can be treated with aggressive multimodality strategies, while those with poor prognostic guess can be offered supportive care. The phase III randomized QUARTZ trial represents an excellent example in this setting, which exhibited no viable advantage of WBRT over dexamethasone plus best supportive care in poor prognostic patients [41]. Therefore, the BM scoring systems may likewise serve as beneficial by the provision of realistic desires to the patients and their caregivers and in properly adjusting the treatment costs [42]. Furthermore, the prognostic scoring frameworks might be of principal significance by stratifying the BM patients with the comparative prognoses in the similar arms of the randomized trials, and in this way, might minimize the confounding factors and meaningfully improve the academic legitimacy of the published results of such investigations in an increasingly trustable manner.

Besides their practical usefulness, all attainable BM scoring systems, unfortunately, have some inherent limitations. One essential common hindrance of every unique prognostic system is the incorporation of relatively more favorable patients’ groups, rendering it troublesome to decide the ideal treatment for patients with comparable unfavorable prognosticators. Additionally, relatively higher accumulation of BM patients in the better score groups brings the question of whether there is a general inclination for the intentional omission of some potentially effective treatment measures in more inferior prognostic groups. This is justifiable somewhat since the best supportive care remains the more frequently chosen management strategy for such patients, but still, this undoubtedly creates unavoidable statistical power bias in head to head comparisons.

Another common drawback is the utility of remarkably diverse prognostic factors in different scoring systems. For instance, the frequently referred RPA and BSBM do not esteem the number of BM as a significant prognostic factor. In the same way, the volume of BM, steroid response and primary tumor site are included merely in the SIR, Rotterdam, and DS-GPA systems, respectively. Moreover, the respective prerequisites for the BM volume for SIR and tedious clinical workup for steroid response evaluation for Rotterdam scores may severely limit their routine usage only for SRS and low-volume radiation oncology clinics.

Though the primary tumor characteristics and the driver mutations were comprehensively addressed in the initial GPA and DS-GPA, yet they didn't include the highly active targeted agents or novel immunotherapies in the scoring systems. However, these novel therapeutic agents may subtly alter the bleak prognosis of patients in a significant manner in their ways. Amply supporting this real-world experience, lapatinib and alectinib have exhibited remarkable clinical activities on BMs from HER-2 positive breast cancer patients and anaplastic lymphoma kinase (ALK) rearranged non-small cell lung cancer patients, separately [43, 44]. Likewise, dabrafenib and vemurafenib demonstrated considerable activity against the BRAF mutated malignant melanoma BMs [42, 45]. Establishing these rational anticipations, Johung et al. assessed the role of driver mutation genotype in predicting recurrence among 496 NSCLC BM patients treated with SRS and revealed that none of the patients with EGFR mutation and EML4-ALK translocation experienced in SRS field relapses [46]. Conversely, 18% of patients with KRAS mutation and 19% without these mutations had in SRS field relapses. Survival analysis was unperformed in this critical investigation; however yet, announced discoveries are significant for the way that they provided valuable insights into the profound influence of driver mutations on radiation efficacy by powerfully suggesting higher radiosensitivity for EGFR and EML4-ALK mutant tumors and relative radioresistance for KRAS mutation-positive BMs.

An impressive study reported by Spanberger et al. assessed the prognostic value of the Ki-67 index, hypoxia-induced factor 1α expression, peritumoral edema, and micro vascularization patterns in 219 patients who underwent neurosurgical resection for BMs [47]. This comprehensive examination is notable for the direct assessment of extra factors for their prognostic worth notwithstanding the entrenched GPA. Other than asserting the legitimacy of GPA, peritumoral edema seemed, by all accounts, to be the unique variable related to prognostic worth. The MRI characteristics were further studied in 65 patients with single BM for their predictive power on survival outcomes, and the preoperative diffusion-weighted imaging signal intensities were found to be significantly correlated with survival results [48]. In another study of 69 BM patients from non-small-cell lung cancer caveolin-1 was surveyed for its predictive role on survival and radiotherapy responsiveness as a pathologic marker [49]. The study outcomes convincingly demonstrated that caveolin-1 expressing BMs were linked with notable more dismal prognoses and an increased risk of death (p=0.015). Moreover, in patients <54 years caveolin-1 expression was shown to neutralize the favorable effect of young age on survival. Unequivocally certifying the presence of a caveolin-1 related radioresistant strain, an increased risk of death was detected in the group with caveolin-1 expressing BMs among the RT receivers (HR=6.839; P=0.004) contrasted with their non-expressing counterparts. Though further corroborative investigations are required, accessible proof humbly proposes that peritumoral edema, diffusion-weighted imaging signal intensities, and caveolin-1 deserve to be investigated for their prognostic worth in future prognostic scoring systems.

Finally, another common impediment of all BM scoring systems is that all variables are merely inferred to foresee OS outcomes and with none focusing on other endpoints, such as time to neurologic progression/decline or BM-specific survival. In affirmation, even the two externally validated famous nomograms reported by Ahn et al. for breast cancer patients and Zindler et al. for non-small-cell lung cancer patients presenting with BMs likewise chose the OS as the essential endpoint [50, 51]. Given that the more frequent use of targeted agents and immunotherapies may positively enhance the intra- and extracranial tumor control rates and the WBRT results in different local control than SRS, it is imperative to control for the confounding effect of a particular treatment strategy to optimally predict outcomes. In this peculiar manner, subsequent investigations ought to be explicitly designed with the ultimate objective of developing novel prognostic models or nomograms which adequately address these particular issues in patients presenting with BMs.

Conclusions

Although too much work is needed to be done to improve the currently available prognostic scoring systems for patients presenting with BMs, yet, they are however worthwhile for true prognostic laddering and most fit treatment arrangements of such patients. However, the results of statistically well-powered further studies addressing the potential prognostic worth of novel molecular, genetic, pathological, radiological, and treatment-related factors are eagerly awaited. Such extra and profoundly explicit markers may hopefully further enhance the prognostic and predictive strength of the entrenched prognostic scoring systems, particularly in the era of targeted and immune-therapies and progressively favored SRS.

Competing Interests

None.

Funding

None.

Acknowledgements

None.

Author Contributions

All authors contributed equally.

Article Info

Article Type

Research Article

Publication history

Received: Thu 20, Feb 2020
Accepted: Mon 09, Mar 2020
Published: Tue 17, Mar 2020

Copyright

© 2023 Erkan Topkan. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Hosting by Science Repository.
DOI: 10.31487/j.COR.2020.03.03

Author Info

Ahmet Kucuk Berrin Pehlivan Duygu Sezen Erkan Topkan Esma Didem Ikiz Eyub Yasar Aydemir Nulifer Kılıc Durankus Sukran Senyurek Ugur Selek Yasemin Bolukbasi

Corresponding Author

Erkan Topkan
Baskent University Medical Faculty, Department of Radiation Oncology, Adana, Turkey

Figures & Tables

Table 1: Prognostic factors for patients presenting with brain metastases.

Factor	Tumor site
Age	Common
Gender	Common
Performance status	Common
Location of BM	Common
Number of BM	Common
Size of BM	Common
BM velocity	Common
Volume of BM	Common
Neurologic deficit status	Common
Extracranial disease status	Common
Histology of BM	Radioresistant vs. radiosensitive
ER/PR status	Breast
HER-2 status	Breast
EGFR status	Non-small cell lung
EML4-ALK status	Non-small cell lung
BRAF status	Malignant melanoma
Caveolin-1	Non-small cell lung
Peritumoral edema status	Common
Radiologic features	Common
Interval from primary diagnosis	Common
Radiotherapy technique (SRS vs. others)	Common
Type of systemic therapy	Non-small cell lung, breast, malignant melanoma

BM: Brain metastasis; ER: Estrogen receptor; PR: Progesteron receptor; HER-2: Human epidermal growth factor receptor 2; EML4-ALK: Echinoderm microtubule associated protein-like 4 and anaplastic lymphoma kinase; BRAF: v-Raf murine sarcoma viral oncogene homolog B.

Table 2: Comparison of frequently utilized prognostic scoring systems for brain metastases.

	RPA (n=1200)	SIR (n=65)	Rotterdam (n=1292)	BSBM (n=110)	Rades (n=1797)	GGS (n=479)	GPA (n=1960)	DS-GPA (n=4259)
Age	Yes	Yes			Yes	Yes	Yes	Yes
Performance status	KPS	KPS	ECOG	KPS	KPS	KPS	KPS	KPS
Primary control	Yes	Yes		Yes
ECM	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Number of BM		Yes			Yes	Yes	Yes	Yes
Volume of BM		Yes
Steroid response			Yes
Tumor histology						Yes	Yes	Yes
Interval from diagnosis to RT					Yes
ER/PR status								Yes
HER-2 status								Yes
EGFR status								Yes
EML4-ALK status								Yes
BRAF status								Yes
AFP								Yes
Child-Pugh-Score								Yes

Note: Empty spaces represents for ‘No’.

RPA: Recursive Partitioning Analysis Scoring System; SIR: Score index for radiosurgery; BSBM: Basic score for brain metastases; GGS: Golden grading system; GPA: Graded prognostic assessment; DS-GPA: Disease specific graded prognostic assessment; ECM: Extracellular matrix; BM: Brain metastasis; RT: Radiotherapy; ER: Estrogen receptor; PR: Progesteron receptor; HER-2: Human epidermal growth factor receptor 2; EGFR: Epidermal growth factor receptor; EML4-ALK: Echinoderm microtubule associated protein-like 4 and anaplastic lymphoma kinase; BRAF: v-Raf murine sarcoma viral oncogene homolog B; AFP: Alpha fetoprotein.

References

1. Nabors LB, Portnow J, Ammirati M, Baehring J, Brem H et al. (2017) NCCN Guidelines Insights: Central Nervous System Cancers, Version 1.2017. J Natl Comprehensive Cancer Netw 15: 1331-1345. [Crossref]
2. Pruitt AA (2017) Epidemiology, Treatment, and Complications of Central Nervous System Metastases. Continuum (Minneap Minn) 23: 1580-1600. [Crossref]
3. Sorensen JB, Hansen HH, Hansen M, Dombernowsky P (1988) Brain metastases in adenocarcinoma of the lung: frequency, risk groups, and prognosis. J Clin Oncol 6: 1474-1480. [Crossref]
4. Kocher M, Wittig A, Piroth MD, Treuer H, Seegenschmiedt H et al. (2014) Stereotactic radiosurgery for treatment of brain metastases. A report of the DEGRO Working Group on Stereotactic Radiotherapy. Strahlenther Onkol 190: 521-532. [Crossref]
5. Nathoo N, Chahlavi A, Barnett GH, Toms SA (2005) Pathobiology of brain metastasis. J Clin Pathol 58: 237-242. [Crossref]
6. Fidler IJ (2015) The biology of Brain Metastasis: Challenges for Therapy. Cancer J 21: 284-293. [Crossref]
7. Landis SH, Murray T, Bolden S, Wingo PA (1998) Cancer statistics, 1998. CA Cancer J Clin 48: 6-29. [Crossref]
8. Norden AD, Wen PY, Kesari S (2005) Brain metastases. Curr Opin Neurol 18: 654-661. [Crossref]
9. Soffietti R, Abacioglu U, Baumert B, Combs SE, Kinhult S et al. (2017) Diagnosis and treatment of brain metastases from solid tumors: guidelines from the European Association of Neuro-Oncology (EANO). Neuro Oncol 19: 162-174. [Crossref]
10. Eichler AF, Loeffler JS (2007) Multidisciplinary management of brain metastases. Oncologist 12: 884-898. [Crossref]
11. Delattre JY, Krol G, Thaler HT, Posner JB (1988) Distribution of brain metastases. Arch Neurol 45: 741-744. [Crossref]
12. Eichler AF, Chung E, Kodack DP, Loeffler JS, Fukumura D et al. (2011) The biology of brain metastases-translation to new therapies. Nat Rev Clin Oncol 8: 344-356. [Crossref]
13. Sperduto PW, Kased N, Roberge D, Xu Z, Shanley R et al. (2012) Summary report on the graded prognostic assessment: an accurate and facile diagnosis-specific tool to estimate survival for patients with brain metastases. J Clin Oncol 30: 419-425. [Crossref]
14. Sperduto PW, Yang TJ, Beal K, Pan H, Brown PD et al. (2017) Estimating survival in patients with lung cancer and brain metastases: an update of the graded prognostic assessment for lung cancer using molecular markers (lung-molGPA). JAMA Oncol 3: 827-831. [Crossref]
15. Sperduto PW, JiangW, Brown PD, Braunstein S, Sneed P et al. (2017) The prognostic value of BRAF, C-KIT, and NRAS Mutations in Melanoma Patients With Brain Metastases. Int J Radiat Oncol Biol Phys 98: 1069-1077. [Crossref]
16. Sperduto PW, Kased N, Roberge D, Xu Z, Shanley R et al. (2012) Effect of tumor subtype on survival and the graded prognostic assessmentfor patients with breast cancer and brain metastases. Int J Radiat Oncol Biol Phys 82: 2111-2117. [Crossref]
17. Lim S, Lee S, Lim JY, Park JS, Seong JS et al. (2014) Hepatocellular carcinoma specific graded prognostic assessment can predict outcomes for patients with brain metastases from hepatocellular carcinoma. J Neurooncol 120: 199-207. [Crossref]
18. Billing PS, Miller DL, Allen MS, Deschamps C, Trastek VF et al. (2001) Surgical treatment of primary lung cancer with synchronous brain metastases. J Thorac Cardiovasc Surg 122: 548-553. [Crossref]
19. Yates JW, Chalmer B, McKegney FP. (1980) Evaluation of patients with advanced cancer using the Karnofsky performance status. Cancer 45: 2220-2224. [Crossref]
20. Abrahams JM, Torchia M, Putt M, Kaiser LR, Judy KD (2001) Risk factors affecting survival after brain metastases from non-small cell lung carcinoma: a follow-up study of 70 patients. J Neurosurg 95: 595-600. [Crossref]
21. Venur VA, Ahluwalia MS (2015) Prognostic scores for brain metastasis patients: use in clinical practice and trial design. Chin Clin Oncol 4: 18. [Crossref]
22. Gaspar L, Scott C, Rotman M, Asbell S, Phillips T et al. (1997) Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials. Int J Radiat Oncol Biol Phys 37: 745-751. [Crossref]
23. Komarnicky LT, Phillips TL, Martz K, Asbell S, Isaacson S et al. (1991) A randomized phase III protocol for the evaluation of misonidazole combined with radiation in the treatment of patients with brain metastases (RTOG-7916). Int J Radiat Oncol Biol Phys 20: 53-58. [Crossref]
24. Phillips TL, Scott CB, Leibel SA, Rotman M, Weigensberg IJ (1995) Results of a randomized comparison of radiotherapy and bromodeoxyuridine with radiotherapy alone for brain metastases: report of RTOG trial 89-05. Int J Radiat Oncol Biol Phys 33: 339-348. [Crossref]
25. Sause WT, Crowley JJ, Morantz R, Rotman M, Mowry PA et al. (1990) Solitary brain metastasis: results of an RTOG/SWOG protocol evaluation surgery + RT versus RT alone. Am J Clin Oncol 13: 427-432. [Crossref]
26. Nieder C, Andratschke N, Grosu AL, Molls M (2003) Recursive partitioning analysis (RPA) class does not predict survival in patients with four or more brain metastases. Strahlenther Onkol 179: 16-20. [Crossref]
27. Weltman E, Salvajoli JV, Brandt RA, de Morais Hanriot R, Prisco FE et al. (2000) Radiosurgery for brain metastases: a score index for predicting prognosis. Int J Radiat Oncol Biol Phys 46: 1155-1161. [Crossref]
28. Lagerwaard FJ, Levendag PC, Nowak PJ, Eijkenboom WM, Hanssens PE et al. (1999) Identification of prognostic factors in patients with brain metastases: a review of 1292 patients. Int J Radiat Oncol Biol Phys 43: 795-803. [Crossref]
29. Lorenzoni J, Devriendt D, Massager N, David P, Ruíz S et al. (2004) Radiosurgery for treatment of brain metastases: estimation of patient eligibility using three stratification systems. Int J Radiat Oncol Biol Phys 60: 218-224. [Crossref]
30. Rades D, Dziggel L, Haatanen T, Veninga T, Lohynska R et al. (2011) Scoring systems to estimate intracerebral control and survival rates of patients irradiated for brain metastases. Int J Radiat Oncol Biol Phys 80: 1122-1127. [Crossref]
31. Andrews DW, Scott CB, Sperduto PW, Flanders AE, Gaspar LE et al. (2004) Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 363: 1665-1672. [Crossref]
32. Sperduto PW, Berkey B, Gaspar LE, Mehta M, Curran W (2008) A new prognostic index and comparison to three other indices for patients with brain metastases: an analysis of 1,960 patients in the RTOG database. Int J Radiat Oncol Biol Phys 70: 510-514. [Crossref]
33. Antoni D, Noel G (2013) [Radiotherapy of brain metastases according to the GPA score (Graded Prognostic Assessment)]. Cancer Radiother 17: 424-427. [Crossref]
34. Tabouret E, Metellus P, Goncalves A, Esterni B, Charaffe Jauffret E et al. (2014) Assessment of prognostic scores in brain metastases from breast cancer. Neuro Oncology 16: 421-428. [Crossref]
35. Luo J, Zhu H, Tang Y, Wang H, Zhou X et al. (2014) Analysis of prognostic factors and comparison of prognostic index scores in patients with brain metastases after whole-brain radiotherapy. Int J Clin Exp Med 7: 5217-5225. [Crossref]
36. Golden DW, Lamborn KR, McDermott MW, Kunwar S, Wara WM et al. (2008) Prognostic factors and grading systems for overall survival in patients treated with radiosurgery for brain metastases: variation by primary site. J Neurosurg 109: 77-86. [Crossref]
37. Sperduto PW, Chao ST, Sneed PK, Luo X, Suh J et al. (2010) Diagnosis-specific prognostic factors, indexes, and treatment outcomes for patients with newly diagnosed brain metastases: a multi-institutional analysis of 4,259 patients. Int J Radiat Oncol Biol Phys 77: 655-661. [Crossref]
38. Sperduto PW, Deegan BJ, Li J, Jethwa KR, Brown PD et al. (2018) Estimating survival for renal cell carcinoma patients with brain metastases: an update of the Renal Graded Prognostic Assessment tool. Neuro Oncol 20: 1652-1660. [Crossref]
39. Lim S, Lee S, Lim JY, Park JS, Seong JS et al. (2014) Hepatocellular carcinoma specific graded prognostic assessment can predict outcomes for patients with brain metastases from hepatocellular carcinoma. J Neurooncol 120: 199-207. [Crossref]
40. Rades D, Dziggel L, Haatanen T, Veninga T, Lohynska R et al. (2011) Scoring systems to estimate intracerebral control and survival rates of patients irradiated for brain metastases. Int J Radiat Oncol Biol Phys 80: 1122-1127. [Crossref]
41. Mulvenna P, Nankivell M, Barton R, Faivre Finn C4, Wilson P et al. (2016) Dexamethasone and supportive care with or without whole brain radiotherapy in treating patients with non-small cell lung cancer with brain metastases unsuitable for resection or stereotactic radiotherapy (QUARTZ): results from a phase 3, non-inferiority, randomised trial. Lancet 388: 2004-2014. [Crossref]
42. McArthur GA, Maio M, Arance A, Nathan P, Blank C et al. (2017) Vemurafenib in metastatic melanoma patients with brain metastases: an open-label, single-arm, phase 2, multicentre study. Ann Oncol 28: 634-641. [Crossref]
43. Petrelli F, Ghidini M, Lonati V, Tomasello G, Borgonovo K et al. (2017)The efficacy of lapatinib and capecitabine in HER-2 positive breast cancer with brain metastases: A systematic review and pooled analysis. Eur J Cancer 84: 141-148. [Crossref]
44. Gadgeel SM, Gandhi L, Riely GJ, Chiappori AA, West HLet al. (2014) Safety and activity of alectinib against systemic disease and brain metastases in patients with crizotinib-resistant ALK-rearranged non-small-cell lung cancer (AF-002JG): results from the dose-finding portion of a phase 1/2 study. Lancet Oncol 15: 1119-1128. [Crossref]
45. Davies MA, Saiag P, Robert C, Grob JJ, Flaherty KT et al. (2017) Dabrafenib plus trametinib in patients with BRAF^V600-mutant melanoma brain metastases (COMBI-MB): a multicentre, multicohort, open-label, phase 2 trial. Lancet Oncol 18: 863-873. [Crossref]
46. Johung KL, Yao X, Li F, Yu JB, Gettinger SN et al. (2013) A clinical model for identifying radiosensitive tumor genotypes in non-small cell lung cancer. Clin Cancer Res 19: 5523-5532. [Crossref]
47. Spanberger T, Berghoff AS, Dinhof C, Ilhan Mutlu A, Magerle M et al. (2013) Extent of peritumoral brain edema correlates with prognosis, tumoral growth pattern, HIF1a expression and angiogenic activity in patients with single brain metastases. Clin Exp Metastasis 30: 357-368. [Crossref]
48. Berghoff AS, Spanberger T, Ilhan MutluA, Magerle M, Hutterer M et al. (2013) Preoperative diffusion-weighted imaging of single brain metastases correlates with patient survival times. PLoS One 8: e55464. [Crossref]
49. Duregon E, Senetta R, Pittaro A, Verdun di Cantogno L, Stella G et al. (2015) CAVEOLIN-1 expression in brain metastasis from lung cancer predicts worse outcome and radioresistance, irrespective of tumor histotype. Oncotarget 6: 29626-29636. [Crossref]
50. Ahn HK, Lee S, Park YH, Sohn JH, Jo JC et al. (2012) Prediction of outcomes for patients with brain parenchymal metastases from breast cancer (BC): a new BC-specific prognostic model and a nomogram. Neuro Oncol 14: 1105-1113. [Crossref]
51. Zindler JD, Jochems A, Lagerwaard FJ, Beumer R, Troost EGC et al. (2017) Individualized early death and long-term survival prediction after stereotactic radiosurgery for brain metastases of non-small cell lung cancer: Two externally validated nomograms. Radiother Oncol 123: 189-194. [Crossref]