**2.5 Literature searches, inclusion and exclusion criteria**

Literature searches are performed in a semi-automated manner, similar to a systematic review (see Cochrane Review instructions for a full description of this method). Here, we utilize keyword searches in PubMed. Our strategy is not to be over-limiting in our resulting study selection. Rather, we limit the size of our dynamic meta-analysis by decreasing the study scope instead of using highly selective inclusion/exclusion criteria.

For the dynamic meta-analysis presented in this chapter, we perform two different literature searches, Phase I and Phase II. The first literature search (Phase I) is an all-encompassing literature search for primary research studies/articles. Phase I inclusion results in ~1,803 papers, while Phase I exclusion leaves a remainder of 1,144 papers. These 1,144 articles are the studies/data sets for dynamic meta-analysis. The second literature search (Phase II) is for ALS review articles. Phase II inclusion results in ~200 reviews, while Phase II exclusion reduces the number to 52. The review articles are utilized for the purposes of structure and aggregation (to be discussed in the following sub-sections).

Phase I Inclusion Criteria:

62 Amyotrophic Lateral Sclerosis

time is introduced explicitly. Therefore, dynamic meta-analysis can incorporate experimental data from differing time-points and predict effects over time. These traits make dynamic meta-analysis unique, even when compared to advanced meta-analysis methods such as network analysis (Trelle, Reichenbach et al. 2011). While network analysis does use comparative relationships, it does not include interactions or show how relationship ratios change over time. Thus, where traditional or even advanced metaanalysis produces a static set of linear relationships, dynamic meta-analysis produces a set of differential equations. This results in an innovative way to examine pathology dynamics as we can look at how metrics change and interact over time rather than being limited to how they correlate at a single point in time. Currently, the only other available technique capable of implictly including interactions and explicity including time is relational modeling (Mitchell, 2009; Mitchell and Lee, 2008). In fact, dynamic meta-analysis is, itself, one form of relational modeling. However, traditional relational models typically do not provide the desired statistical weight of dynamic meta-analysis since only one primary

Dynamic meta-analysis is similar to traditional meta-analysis in that it utilizes literature searches, inclusion/exclusion critieria, and data aggregation techniques. A key difference in the dynamic meta-analysis process, however, is the study structure and data extraction. In the following sub-sections of the chapter we provide the details necessary to perform each step of dynamic meta-analysis: determining the study scope, performing literature searches and study inclusion/excusion, developing structure, extracting data, aggregating extracted data, implementing dynamic meta-analysis, and analyzing dynamic meta-analysis results.

Just as in traditional meta-analysis, defining the scope is an important step. There are several things to consider, including the outcome goals of the project, the measures and timepoints to be included, the statistical weight, and the desired timeline of the project. There is no methodological limit on the number of studies, measures, and timepoints that can be included in dynamic meta-analysis. Rather, the researcher must impose those limits. There is a balance between including enough studies to obtain statistically significant results and the amount of man-hours it takes to perform dynamic meta-analysis. The one drawback of meta-analysis is it is by no means a completely automated process. Rather, humans must be involved at each step, to search and more importantly extract the data from included

To assist in balancing workload and the time it takes to get preliminary results from dynamic meta-analysis, we divide dynamic meta-analysis into two parts: a feasibility study and a full study. A full study, as the name implies, encompasses all of the primary articles that meet the inclusion criteria. In contrast, a feasibility study can potentially have the same number and breadth of outcome measures as the full study, but utilizes a lesser number of included primary studies for each metric (i.e. a lower "n"). That is, the statistical weight is decreased. The advantage of performing an initial feasibility study goes beyond simply obtaining preliminary results more quickly. The initial results of the feasibility study also provide insight and direction, which can be used to fine tune the targeting of the full study (e.g. determining if

more or less measures are needed or if the scope of the study needs revised, etc.).

study is included per interaction.

**2.4 Defining the study scope** 

studies.

**2.3 Overview of the dynamic meta-analysis process** 


Phase I Exclusion Criteria:


Phase II Inclusion Criteria:


Phase II Exclusion Criteria:


### **2.6 Structure: Outcome measures, interactions, and categories**

Similar to traditional meta-analysis, dynamic meta-analysis utilizes outcome measures (measures that are calculated or derived from the included studies) as a means of prediction. Unlike traditional meta-analysis, dynamic meta-analysis also includes *interactions* between the outcome measures. However, due to the larger scope of dynamic meta-analysis compared to traditional meta-analysis, it is helpful to combine individual experimental metrics or outcome metrics into aggregates we refer to as *categories*. The outcome measures, interactions, and their respective categories, together, make up the dynamic meta-analysis structure (Figure 1).

Dynamic Meta-Analysis as a Therapeutic Prediction Tool for Amyotrophic Lateral Sclerosis 65

follow at the end of this sub-section): axonal transport, chemistry, energetics, excitotoxicity, free radicals, genetics, inflammation, necro-apoptosis, proteomics, and systemics. Reviews were also used to preliminarily determine which categories are inter-related (without having to examine each and every primary article). This is extremely helpful for development of the feasibility study. The quantitative specifics of the data aggregation process are discussed in Data Aggregation, while the category definitions are given below. *Excitotoxicity*: encompasses measures of electrophysiology; ion, neurotransmitter, and buffer concentrations; activation of ionotropic/metabotropic receptors (Ikonomidou, Qin Qin et al. 1996; Dunlop, Beal McIlvain et al. 2003; Van Damme, Leyssen et al. 2003); altered excitability related to sodium (Kuo, Schonewille et al. 2004; Kuo, Siddique et al. 2005); and transport

and pump capacity (Guatteo, Carunchio et al. 2007), causing toxic over-activation.

(Ackerley, Grierson et al. 2004).

*Axonal transport*: encompasses measures of the anterograde and retrograde transport of cargos, such as mitochondria, neurotransmitters, neurofilaments, and endosomes/ lysosomes, as well as the measures of the involved machinery, such as the molecular motors kinesin and dynein. The most recognized impairments and their measures include mutations to cargos (Meyer and Potter 1995; Wong, He et al. 2000) and molecular motor cargo carriers (Hafezparast, Ahmad-Annuar et al. 2003; Teuchert, Fischer et al. 2006; Mitchell and Lee, 2009) that prevent the cargo from be appropriately bound to either dynein or kinesin. Other deficits include correlations to energetics, such as decreased mitochondrial transport or a decrease in overall transport due to a drop in mitochondrial potential

*Energetics:* encompasses measures of all machinery and processes related to cellular respiration and production of cellular ATP (Echaniz-Laguna, Zoll et al. 2002). Energetic contributors include impairments to the cellular machinery responsible for the production of ATP, especially mitochondria (Kong and Xu 1998; Sumi, Nagano et al. 2006), whose dysfunction also leads to accumulation of free radicals and calcium. Overburdened energetic capabilities because of increased homeostatic and transport demands from

*Genetic damage:* encompasses measures of an extremely diverse spectrum of either inherited or sporadic mutations resulting in cellular dysfunction (Tanaka, Niwa et al. 2006). The most widely known genetic mutation, accounting for 2% of all ALS cases, is superoxide dismutase-1 (SOD1), which has over 100 known different mutations (Banci, Bertini et al. 2008) (for list, see www.alsod.org) that result in a gain of one or more toxic properties that

*Proteomics*: encompasses measures of protein folding, degradation, and translation, which become impaired, resulting in defective essential proteins, toxic accumulation of aggregates or aggresomes, and inhibition of organelle function (Watanabe, Dykes-Hoberg et al. 2001;

*Chemistry*: encompasses measures of aberrant cellular chemistry, enzymatics, or catalysis (Tiwari, Xu et al. 2005; Tokuda, Ono et al. 2008) that results in oxidative damage and metal

*Inflammation*: encompasses measures of immune-induced inflammation, including astrocyte (Nagai, Re et al. 2007) and microglia (Hall, Andrus et al. 1998) counts, gliosis and the release of nitric oxide and proinflammatory cytokines, which in combination with impaired growth factors/trophic support (Narai, Nagano et al. 2005; Kadoyama, Funakoshi et al. 2007),

are independent of the levels of SOD1 activity (Stathopulos, Rumfeldt et al. 2003).

mishandling that can be seen alone or in conjunction with SOD1 mutations.

excitotoxicity (Dupuis, Oudart et al. 2004) have also been observed.

Urushitani, Kurisu et al. 2002; Rumfeldt, Lepock et al. 2009).

further inhibit the neural environment.

Fig. 1. Dynamic Meta-Analysis Feasibility Study Structure. Boxes represent the ten categories of outcome measures. Lines represent the 72 one-way interactions between categories of outcome measures. Systemics is shown in red as it represents the functional outcome metrics measured the in vivo G93A SOD1 mouse, and is used as the primary outcome for treatment evaluation (see Results). Categories are derived from the Phase II literature reviews. The outcome measures and their interactions are derived from the primary studies obtained from the Phase I literature search.

### **2.6.1 Category definitions**

Aggregation of primary study metrics into categories balances the number and specificity of dynamic meta-analysis outcome measures with statistical weight. Too little aggregation will result in too disparate of a collection of very specific experimental mesasures or outcomes and will reduce the statistical weight of their predictions. In contrast, too much aggregation will result in outcome measures that are too broad; while this will increase their stasticial weight, it will also ultimately reduce the specificity of the dynamic meta-analysis predictions. Thus, the level, type, and implementation of aggregation will depend on the scope of the study, the number or primary studies utilized, as well as the desired statistical weight.

If the quantitative outcome metrics and their measured interactions come from the primary studies of Phase I, where do their categories come from? Categories are derived from the review articles of Phase II. Reviews do a nice job of providing key topics that are being researched by the field. For this feasibility study, reviews were analyzed for common research topics, based on broad categories of related physiological measures. For example, all measures of dynein, kinesin, mitochondrial transport, neurofilament transport, and neurofilament transport, etc, were grouped into the category outcome measure "Axonal Transport". Key terms were then extracted from these topics and used to sort the primary studies, and their respective outcome measures, into the following categories (definitions

Fig. 1. Dynamic Meta-Analysis Feasibility Study Structure. Boxes represent the ten categories of outcome measures. Lines represent the 72 one-way interactions between categories of outcome measures. Systemics is shown in red as it represents the functional outcome metrics measured the in vivo G93A SOD1 mouse, and is used as the primary outcome for treatment evaluation (see Results). Categories are derived from the Phase II literature reviews. The outcome measures and their interactions are derived from the

Aggregation of primary study metrics into categories balances the number and specificity of dynamic meta-analysis outcome measures with statistical weight. Too little aggregation will result in too disparate of a collection of very specific experimental mesasures or outcomes and will reduce the statistical weight of their predictions. In contrast, too much aggregation will result in outcome measures that are too broad; while this will increase their stasticial weight, it will also ultimately reduce the specificity of the dynamic meta-analysis predictions. Thus, the level, type, and implementation of aggregation will depend on the scope of the study, the number or primary studies utilized, as well as the desired statistical

If the quantitative outcome metrics and their measured interactions come from the primary studies of Phase I, where do their categories come from? Categories are derived from the review articles of Phase II. Reviews do a nice job of providing key topics that are being researched by the field. For this feasibility study, reviews were analyzed for common research topics, based on broad categories of related physiological measures. For example, all measures of dynein, kinesin, mitochondrial transport, neurofilament transport, and neurofilament transport, etc, were grouped into the category outcome measure "Axonal Transport". Key terms were then extracted from these topics and used to sort the primary studies, and their respective outcome measures, into the following categories (definitions

primary studies obtained from the Phase I literature search.

**2.6.1 Category definitions** 

weight.

follow at the end of this sub-section): axonal transport, chemistry, energetics, excitotoxicity, free radicals, genetics, inflammation, necro-apoptosis, proteomics, and systemics. Reviews were also used to preliminarily determine which categories are inter-related (without having to examine each and every primary article). This is extremely helpful for development of the feasibility study. The quantitative specifics of the data aggregation process are discussed in Data Aggregation, while the category definitions are given below.

*Excitotoxicity*: encompasses measures of electrophysiology; ion, neurotransmitter, and buffer concentrations; activation of ionotropic/metabotropic receptors (Ikonomidou, Qin Qin et al. 1996; Dunlop, Beal McIlvain et al. 2003; Van Damme, Leyssen et al. 2003); altered excitability related to sodium (Kuo, Schonewille et al. 2004; Kuo, Siddique et al. 2005); and transport and pump capacity (Guatteo, Carunchio et al. 2007), causing toxic over-activation.

*Axonal transport*: encompasses measures of the anterograde and retrograde transport of cargos, such as mitochondria, neurotransmitters, neurofilaments, and endosomes/ lysosomes, as well as the measures of the involved machinery, such as the molecular motors kinesin and dynein. The most recognized impairments and their measures include mutations to cargos (Meyer and Potter 1995; Wong, He et al. 2000) and molecular motor cargo carriers (Hafezparast, Ahmad-Annuar et al. 2003; Teuchert, Fischer et al. 2006; Mitchell and Lee, 2009) that prevent the cargo from be appropriately bound to either dynein or kinesin. Other deficits include correlations to energetics, such as decreased mitochondrial transport or a decrease in overall transport due to a drop in mitochondrial potential (Ackerley, Grierson et al. 2004).

*Energetics:* encompasses measures of all machinery and processes related to cellular respiration and production of cellular ATP (Echaniz-Laguna, Zoll et al. 2002). Energetic contributors include impairments to the cellular machinery responsible for the production of ATP, especially mitochondria (Kong and Xu 1998; Sumi, Nagano et al. 2006), whose dysfunction also leads to accumulation of free radicals and calcium. Overburdened energetic capabilities because of increased homeostatic and transport demands from excitotoxicity (Dupuis, Oudart et al. 2004) have also been observed.

*Genetic damage:* encompasses measures of an extremely diverse spectrum of either inherited or sporadic mutations resulting in cellular dysfunction (Tanaka, Niwa et al. 2006). The most widely known genetic mutation, accounting for 2% of all ALS cases, is superoxide dismutase-1 (SOD1), which has over 100 known different mutations (Banci, Bertini et al. 2008) (for list, see www.alsod.org) that result in a gain of one or more toxic properties that are independent of the levels of SOD1 activity (Stathopulos, Rumfeldt et al. 2003).

*Proteomics*: encompasses measures of protein folding, degradation, and translation, which become impaired, resulting in defective essential proteins, toxic accumulation of aggregates or aggresomes, and inhibition of organelle function (Watanabe, Dykes-Hoberg et al. 2001; Urushitani, Kurisu et al. 2002; Rumfeldt, Lepock et al. 2009).

*Chemistry*: encompasses measures of aberrant cellular chemistry, enzymatics, or catalysis (Tiwari, Xu et al. 2005; Tokuda, Ono et al. 2008) that results in oxidative damage and metal mishandling that can be seen alone or in conjunction with SOD1 mutations.

*Inflammation*: encompasses measures of immune-induced inflammation, including astrocyte (Nagai, Re et al. 2007) and microglia (Hall, Andrus et al. 1998) counts, gliosis and the release of nitric oxide and proinflammatory cytokines, which in combination with impaired growth factors/trophic support (Narai, Nagano et al. 2005; Kadoyama, Funakoshi et al. 2007), further inhibit the neural environment.

Dynamic Meta-Analysis as a Therapeutic Prediction Tool for Amyotrophic Lateral Sclerosis 67

Therefore, the slope or gain (dY/dX) for wild-type is -0.2 for the interaction from glutamate to retrograde transport. Correspondingly, the gain, for SOD1, is -0.25. Because we only calculate relative changes in dynamic meta-analysis, the magnitude of the gain utilized is actually the relative difference between the wild-type and SOD1 gain magnitudes, (|SOD1| – |wild-type|)/|wild-type|, or [(0.25) – (0.20)]/(0.25) = 0.25, and the sign of this final gain value for this example is net negative (-). Applying the gain of the experimental outcome measures to the category measures, the interaction *from* the category outcome measure of excitotoxicity *to* the category outcome measure of axonal transport is -0.25. Finally, we divide the gain by the time point (or in the case of a feasibility study, the *to* category time

Fig. 2. Scope of dynamic meta-analysis study for the G93A SOD1 mouse model of ALS:

feasibility study versus a full study.

constant) to obtain the interaction coefficient (B) used in Equation 2.

*Free radicals:* encompasses measures of oxidation or inflammation-induced nitric oxide, but particularly the accumulation of reactive oxygen species, such as the superoxides and peroxides that are associated with mitochondrial dysfunction or failure. Free radicals initiate reactions that damage DNA (Pehar, Vargas et al. 2007).

*Necro-apoptosis*: encompasses the measures of cell death, including the signaling cascades, their constituents, and machinery, which promote cell death. The final destination of an ALS affected motoneuron is cell death either through inflammation-induced necrosis or more likely through apoptosis (Mattson and Duan 1999) via the activation of stress response and caspase pathways (Beere 2004; Gifondorwa, Robinson et al. 2007).

*Systemic:* encompasses invivo measures of function in the G93A mouse model (Derave, Van Den Bosch et al. 2003). It includes measures of muscle weakness, atrophy, fasiculations, denervation and ultimately loss of function that decreases essential stimulatory retrograde signaling, causing further progression of the diseased state.
