The genetics of amyotrophic lateral sclerosis

Abstract

Purpose of review

Amyotrophic lateral sclerosis (ALS) has a strong genetic basis, but the genetic landscape of ALS appears to be complex. The purpose of this article is to review recent developments in the genetics of ALS.

Recent findings

Large-scale genetic studies have uncovered more than 40 genes contributing to ALS susceptibility. Both rare variants with variable effect size and more common variants with small effect size have been identified. The most common ALS genes are C9orf72, SOD1, TARDBP and FUS. Some of the causative genes of ALS are shared with frontotemporal dementia, confirming the molecular link between both diseases. Access to diagnostic gene testing for ALS has to improve, as effective gene silencing therapies for some genetic subtypes of ALS are emerging, but there is no consensus about which genes to test for.

Summary

Our knowledge about the genetic basis of ALS has improved and the first effective gene silencing therapies for specific genetic subtypes of ALS are underway. These therapeutic advances underline the need for better access to gene testing for people with ALS. Further research is needed to further map the genetic heterogeneity of ALS and to establish the best strategy for gene testing in a clinical setting.

INTRODUCTION

Amyotrophic lateral sclerosis (ALS) is a heterogeneous neurodegenerative disorder, primarily affecting the motor system [1]. Degeneration of upper motor neurons in the motor cortex and of lower motor neurons in the brainstem and spinal cord give rise to progressive motor impairment, with muscle weakness, wasting and spasticity. Survival is typically limited to 2–5 years after disease onset, due to respiratory muscle weakness. Riluzole and multidisciplinary care with nutritional and ventilatory support remain the cornerstone of the management of ALS [2]. In up to 50% of patients, there is extramotor involvement with cognitive and behavioral impairment, due to frontal and anterior temporal neuronal loss [3]. ALS is a familial disorder in about 10% of patients (fALS); the remaining 90% do not have affected family members and are classified as sporadic ALS (sALS). ALS has a strong genetic basis, and since the discovery of the first ALS gene (superoxide dismutase 1, SOD1) in 1993 [4], more than 40 additional ALS genes have been described [5,6]. In a large proportion of patients with fALS (50–85%), an underlying mendelian gene mutation can be found [7], but also in up to 10–20% of sALS [8^▪,9^▪,10]. Most genes have an autosomal dominant inheritance, but reduced penetrance is a common theme. In addition, rare and common genetic variants with smaller effect size contribute to ALS risk, and combinations of genetic variants occur, suggesting that the genetic landscape of ALS is complex [6]. Although ALS is genetically a very heterogeneous disorder, the disease protein aggregating in up to 97% of patients is TAR DNA-binding protein 43 (TDP-43) [11]. TDP-43 is an RNA and DNA-binding protein, which mostly resides in the nucleus under normal conditions. It is able to shuttle to the cytoplasm and plays important roles in transcription, splicing and RNA transport. In ALS, mislocalization of TDP-43 to the cytoplasm with nuclear depletion and cytoplasmic aggregation, phosphorylation and cleavage is observed in motor neurons and glial cells. How mutations in a diverse set of genes lead to TDP-43 disease is poorly understood and how patients without any gene mutation develop TDP-43 disease remains enigmatic. Two notable exceptions which do not display TDP-43 pathology are ALS caused by mutations in SOD1 and fused in sarcoma (FUS), which are associated with aggregation of the respective proteins. The progress in our understanding of the molecular genetics of ALS has opened avenues for creating disease models to unravel disease pathways leading to motor neuron degeneration. This has resulted in the first FDA- and EMA-approved drug for SOD1-ALS, as intrathecally administered antisense oligonucleotides against SOD1 can reduced neurofilaments (a marker of neuronal/axonal injury) and slow down disease progression [12^▪▪]. A similar gene silencing therapy for FUS-ALS is being developed and the first results look promising [13^▪▪]. These precision medicine approaches are changing the therapeutic landscape of ALS, and a series of other targeted molecular therapies are now under development [14^▪].

Gene testing for ALS using whole exome or whole genome sequencing and gene panel analysis is emerging, but many questions about which genes to include, how to report variants with small or uncertain effect size and how to provide genetic counseling for such variants remain. The scope of this review is to give an overview of recent advancements in the field of genetics of ALS, about genotype-phenotype correlations and how this could affect gene testing in a clinical setting. 

Box 1.

no caption available

TEXT OF REVIEW

Heritability of amyotrophic lateral sclerosis

The heritability of ALS is quite high. In the majority of patients with fALS, an underlying gene mutation can be found, but also in patients with apparently sALS, the heritability is estimated to be 40–60%, based on twin and parent-offspring pair studies [15,16]. Although the genetic predisposition to develop ALS appears to be high, epigenetic factors, aging and environmental exposures also contribute to ALS risk [17]. Part of the heritability of ALS in simplex cases can be explained by monogenetic pathogenic gene mutations in known ALS genes. Extensive gene testing reveals mutations in 10–20% of sALS cases, but the majority of them remains unexplained. A small proportion of ALS risk is attributable to common variants [18]. Rare less damaging variants and combinations of different types of variants most likely contribute as well, but larger studies are required to map them better and combine them in polygenetic risk scores.

Types of amyotrophic lateral sclerosis genes

More than 40 genes have been associated with ALS over the last decades using next-generation sequencing techniques. The heterogeneity in disease causes is overwhelming, but many ALS genes tend to cluster in pathways of protein degradation machinery, RNA metabolism, axonal function and DNA repair [19] (Fig. 1). The genetic variants range from highly penetrant rare variants, over rare variants with moderate effect size to common variants with small effect size, illustrating the complex genetic landscape of ALS (Table 1) [56–80]. The prevalence of different gene mutations is also variable and may differ geographically. An overview of the most common gene mutations is given below. Many of the ALS genes can also cause ALS-FTD or FTD, underscoring the molecular link between both disorders. The reasons why some individuals with such gene mutations develop ALS or rather FTD remain unclear.

FIGURE 1.

Overview of most important ALS-related genes and their roles in motor neurons. (a) NEK1 and C21orf2 play a role in ciliogenesis and cilia maintenance. (b) VCP, TBK1, C9orf72, UBQLN1, SQSTM1, VAPB, CHMP2B, ANXA11, SIGMAR1, OPTN, FIG4, ALS2 and CCNF are involved in the proteasomal and/or autophagy-mediated degradation of misfolded proteins. (c) CHCHD10, SOD1 and TDP-43 are involved in the regulation of mitochondrial dynamics. (d) FUS, TDP-43, NEK1 and C21orf2 facilitate DNA repair. (e) TDP-43, FUS, ANG, SETX, HNRNPA1/A2B1, MATR3, ATXN2 and TIA-1 regulate the RNA metabolism of motor neurons at various levels. (f) ALS2, PFN1, NEFH, PRPH, TUBA4A, SPG11 and NEK1 contribute to the assembly and/or stability of cytoskeletal components. KIF5A and DCTN1 are key players in bidirectional axonal transport. Hh, hedgehog; SMO, Smoothened receptor; UPS, ubiquitin-proteasome system; SG, stress granule; pri-miRNA, primary microRNA; premiRNA, precursor microRNA. Figure created with BioRender.com.

Table 1.

Overview of the most important amyotrophic lateral sclerosis genes

Gene	Chromosomal Location	Inheritance pattern	Segregation in ≥6 families	Frequency fALS	Frequency sALS	Estimated OR	Phenotype	Implicated Pathway
Rare variants with high effect size (OR >10)
C9orf72 [20,21]	9p21.2	AD	Yes	30–60%	5–10%	13.19–35.57	Adult Bulbar onset ALS/FTD	Proteostasis
SOD1 [4]	21q22.11	AD	Yes	12–20%	1–2%	54	Adult Classic LMN dominant	Proteostasis, Oxidative stress
TARDBP [25–27]	1p36.22	AD	Yes	3–5%	1%	24.57	Adult Classic UMN dominant Early onset Slow progression	RNA metabolism, stress granules
FUS [28]	16p11.2	AD, AR	Yes	2–4%	1%	36.19	Adult: Classic Juvenile: fast progression	RNA metabolism, stress granules
TBK1 [29,30]	12q14.2	AD	Yes	3%	<1	52.92	Adult Classic ALS/FTD	Proteostasis
TUBA4A [56]	2q35	AD	/	<1%	<1%	36	Adult Classic	Cytoskeleton
UBQLN2 [57]	Xp11.21	X-linked, AD	/	<1%	<1%	22.11	Adult, Juvenile Bulbar onset ALS/FTD	Proteostasis
VCP [58]	9p13.3	AD	/	1–2%	1%	14.15	Adult Classic ALS/FTD	Proteostasis
OPTN [33]	10p13	AD, AR	Yes	1–4%	<1%	13.62	Adult Spinal onset Slow progression	Proteostasis
NEK1 [34]	4q33	AD	/	1–2%	<1%	11.84	Adult Classic	Cytoskeleton, DNA repair, Cilia
Rare variants with intermediate effect size [1.2<OR<10]
TIA1 [59]	2p13.3	AD	/	2%	<1%	6.9	Adult Classic ALS/FTD	RNA metabolism, stress granules
ATXN2 [39]	12q24.12	AD	/	1–2%	1–3%	6.31	Adult Spinal onset ALS/FTD	RNA metabolism, Stress granule
ANXA11 [60]	10q22.3	AD	Yes	1%	1–2%	4.53	Adult Bulbar onset	Proteostasis
SQSTM1 [61]	5q35.3	AD	/	<1%	<1%	3.91	Adult Classic ALS/FTD	Proteostasis
ANG [62]	14q11.2	AD	/	<1%	<1%	3.77	Adult Classic	RNA metabolism
PRPH [63]	12q13.12	AD, AR	/	/	<1%	3.51	Adult Spinal onset	Cytoskeleton
PFN1 [64]	17p13.2	AD	/	3%	<1%	3.26	Adult Spinal onset Flail leg	Cytoskeleton, Mitochondrial dysfunction
SETX [65]	9qp34.13	AD	/	<1%	<1%	3.19	Juvenile	RNA metabolism, DNA repair
SPG11 [66]	15q21.1	AD	Yes	<1%	<1%	2.88	Juvenile Bulbar/spinal onset Slow progression	Axonal transport
FIG4 [67]	6q21.1	AD, AR	/	<1%	<1%	2.56	Adult Bulbar/spinal onset	Proteostasis
DCTN1 [68]	2p13.1	AD	/	<1%	<1%	2.51	Adult Classic Slow progression	Axonal transport
ALS2 [69]	2q33.1	AD	/	<1%	<1%	2.48	Juvenile UMN dominant	Proteostasis
VAPB [70]	20q13.32	AD	Yes	<1%	<1%	2.04	Adult Spinal LMN dominant	Proteostasis
GLT8D1 [71]	2p21.1	AD	/	<1%	<1%	1.89	Adult Classic	Proteostasis
NEFH [72]	22q12.2	AD	/	<1%	<1%	1.73	Adult Classic	Cytoskeleton
C21orf2/ CFAP410 [18]	21q22.3	NA	/	<1%	<1%	1.65	Adult Classic	Cytoskeleton, DNA repair, Cilia
CHMP2B [73]	3p11.2	AD	/	<1%	<1%	1.61	Adult Bulbar ALS/FTD	Proteostasis
CHCHD10 [74]	22q11.23	AD	/	<1%	<1%	1.51	Adult Bulbar ALS/FTD	Mitochondrial dysfunction
KIF5A [37,38]	12q13.3	AD	Yes	0.5–3%	<1%	1.38	Adult Spinal Early onset Slow progression	Axonal transport
Rare variants with small effect size (OR <1.2)
CCNF [75]	16p13.3	AD	/	1–3%	<1%	1.13	Adult Classic	Proteostasis
Rare variants with unknown effect size
ERBB4 [76]	2q34	AD	/	1%	<1%	NA	Adult Classic	Tyrosine kinase
HNRNPA1 [77]	12q13.13	AD	/	<1%	<1%	NA	Adult Spinal onset Flail arm Slow progression	RNA metabolism
HNRNPA2B1 [77]	7p15.2	AD	/	<1%	<1%	NA	Adult Spinal onset Flail arm	RNA metabolism
MATR3 [78]	5q31.3	AD	Yes	<1%	<1%	NA	Adult Classic	Proteostasis
SIGMAR1 [79]	9p13.3	AD, AR	/	<1%	<1%	NA	Juvenile Spinal onset Slow progression	Proteostasis
SPTLC1 [80]	9q22.31	De novo	/	<1%	<1%	NA	Juvenile Spinal onset LMN dominant	Sphingolipid synthesis
Common variants with small effect size
UNC13A [44]	19p13.11	/	/	/	15%	1.12	Adult Bulbar onset Fast progression	Synaptic function
ERGIC1 [43]	5q35	/	/	/	39.7%	NA	NA	Proteostasis
SLC9A8 [43]	20q13.13	/	/	/	35.3%	NA	NA	Ion homeostasis
SCFD1 [43]	14q12	/	/	/	33.7%	NA	NA	Proteostasis
RPSA [43]	3p21.3	/	/	/	29%	NA	NA	RNA metabolism
COG3 [43]	13q14.13	/	/	/	25.9%	NA	NA	Proteostasis
GPX3 [43]	5q33.1	/	/	/	25.3%	NA	NA	Oxidative stress
PTPRN2 [43]	7q36.3	/	/	/	12.4%	NA	NA	Tyrosine phosphatase
HLA [43]	6p21.3	/	/	/	9.6%	NA	NA	Immune response

ALS2, alsin; ANG, angiogenin; ANXA11, annexin A11; ATXN2, ataxin 2; C21orf2, chromosome 21 open reading frame 2; C9orf72, chromosome 9 open reading frame 72; CCNF, cyclin F; CFAP410, cilia and flagella associated protein 410; CHCHD10, coiled-coil-helix-coiled-coil-helix domain containing 10; CHMP2B, charged multivesicular body protein 2B; COG3, component of oligomeric Golgi complex 3; DCTN1, dynactin subunit 1; ERBB4, Erb-B2 receptor tyrosine kinase 4; ERGIC1, endoplasmic reticulum-Golgi intermediate compartment 1; FIG4, FIG4 phosphoinositide 5-phosphatase; FUS, fused in sarcoma; GLT8D1, glycosyltransferase 8 domain containing 1; GPX3, glutathione peroxidase 3; HNRNPA1, heterogeneous nuclear ribonucleoprotein A1; HNRNPA2B1, heterogeneous nuclear ribonucleoprotein A2/B1; KIF5A, kinesin family member 5A; MATR3, matrin 3; NA, not assessed; NEFH, neurofilament heavy chain; NEK1, NIMA-related kinase 1; OPTN, optineurin; OR, odds ratio; PFN1, profilin 1; PRPH, peripherin; PTPRN2, protein tyrosine phosphatase receptor type N2; RPSA, ribosomal protein SA; SCFD1, Sec1 family domain containing 1; SETX, senataxin; SIGMAR1, sigma nonopioid intracellular receptor 1; SLC9A8, solute carrier family 9 member A8; SOD1, superoxide dismutase 1; SPG11, spastic paraplegia 11; SPTLC1, serine palmitoyltransferase long chain base subunit 1; SQSTM1, sequestosome 1; TARDBP, TAR DNA-binding protein; TBK1, TANK-binding kinase 1; TIA1, cytotoxic granule associated RNA Binding Protein; TUBA4A, tubulin alpha 4A; UBQLN2, ubiquilin 2; UNC13A, Unc-13 homolog A; VAPB, VAMP associated protein B and C; VCP, valosin containing protein.

Rare variants with high effect size

Such pathogenic mutations are found in well established ALS genes and reported to segregate with disease in multiple pedigrees. The five most common ALS genes are chromosome 9 open reading frame 72 (C9orf72), superoxide dismutase 1 (SOD1), TAR DNA-binding protein (TARDBP), fused in sarcoma (FUS) and TANK-binding kinase 1 (TBK1). A reduced penetrance has been described for all these genes, and thus, a negative family history does not exclude the presence of a hereditary form of ALS.

Heterozygous hexanucleotide (GGGGCC) repeat expansions in C9orf72 are by far the most common, responsible for 30–60% of fALS and 5–10% of sALS [20,21]. It is also a common cause of FTD and ALS-FTD. C9orf72 plays a role in endosome maturation and autophagy and both loss-of-function and gain-of-function disease mechanisms have been implicated in the disease. The repeat expansion may hamper transcription of C9orf72, but is also transcribed in sense and antisense RNA, which leads to formation of RNA foci (containing repeat RNA and sequestered RNA-binding proteins) and to translation into dipeptide repeat proteins, which can be toxic [22].

Mutations in SOD1 (mostly dominant missense mutations) explain 10–20% of fALS and 1–2% of sALS. SOD1 gene mutation carriers rarely have concomitant FTD, although frontotemporal involvement has been reported [23,24]. SOD1 is involved in the defense against free radicals, but the mutations cause ALS by inducing SOD1 aggregation and toxicity. Reducing the expression of SOD1 by intrathecal administration of antisense oligonucleotides against SOD1 results in a drastic reduction of neurofilaments, which is followed by a clinical response (with reduced or even halted disease progression, with some improvements in motor function in a proportion of patients) [12^▪▪].

Mutations in TARDBP (mostly heterozygous missense mutations) are responsible for 3–5% of fALS and found in less than 1% of sALS patients [25–27]. Frontotemporal involvement is uncommon in this type of ALS. TARDBP encodes the protein TDP-43 and missense mutations cluster in the C-terminal part of the protein (encoded by exon 6). Mutations render the protein more prone to aggregation with cytoplasmic mislocalization and nuclear clearance as a consequence.

Mutations in FUS (mostly heterozygous missense mutations) cause 2–4% of fALS and less than 1% of sALS [28]. De-novo mutations occur. Some mutations (in particular the P525L mutation) give rise to a very aggressive form of juvenile ALS, while other mutations cause adult-onset ALS, usually without cognitive or behavioral problems. FUS is also an RNA and DNA-binding protein and mutations cluster in the C-terminus of the protein containing the nuclear localization signal. Cytoplasmic aggregation of FUS and nuclear loss-of-function are thought to cause motor neuron loss.

Mutations in TBK1 are found in 1–3% of fALS and in less than 1% of sALS [29,30]. Most mutations are heterozygous loss-of-function mutations, suggesting a haploinsufficiency mechanism. TBK1 is kinase involved in autophagy regulation, for example, by phosphorylating the autophagy receptor optineurin (OPTN) and SMCR8, which forms a complex with C9orf72, thereby promoting autophagosome maturation [31,32].

OPTN mutations are found in 1–4% of fALS and in 0.4% of sALS, in particular is Asian populations [33]. The disease most commonly has a limb onset with long duration before respiratory muscle weakness develops. FTD is not common, but is present in some patients. OPTN is a multifunctional protein, which is also active as autophagy receptor.

Heterozygous loss-of-function or missense mutations in NIMA-related kinase 1 (NEK1) are present in 1–2% of fALS and 0.9% of sALS [34]. Bulbar and upper limb onset is more common and patients display prominent upper motor neuron signs, but there usually is no frontotemporal involvement. NEK1 is a kinase, which phosphorylates its binding partner C21orf2 and this complex has been thought to play a role in DNA damage repair and primary cilia function [35,36].

Rare variants with intermediate effect size

Rare heterozygous missense and loss-of-function variants in C21orf2 were found to be associated with an increased ALS risk, confirming the importance of the NEK1-C21orf2 complex for ALS [18].

Kinesin Family Member 5A (KIF5A) mutations are found in 0.5% of fALS and 0.1% of sALS [37,38]. The mutations cluster in exon 27 and alter the tail domain of the protein. The disease progression is typically somewhat slower and is characterized by prominent upper motor neuron signs. KIF5A is involved in anterograde axonal transport.

Intermediate repeat expansions in the ataxin2 gene (ATXN2) have been associated with ALS [39,40]. ATXN2 was found to be a modifier of TDP-43 toxicity and reducing ATXN2 expression was later shown to modify the phenotype of TDP-43 overexpression mice [41]. Repeat of at least 33 cause spinocerebellar ataxia type 2, but intermediate repeat lengths (in particular 29–33) are associated with an increased risk of developing ALS (odds ratio 3.06) [42]. ATXN2 lowering using ASOs is currently being investigated as therapy for ALS [14^▪].

Common variants with small effect size

Genome-wide association studies have revealed 15 loci associated with ALS risk, but many of them tag rare variants in known ALS genes [43]. The best studied common variant is rs12608932 in Unc13 Homolog A (UNC13A) [44], which confers a small risk for both ALS and FTD [45^▪]. Homozygosity for the C-allele at rs12608932, which is present in about 15% of individuals, also modifies the ALS phenotype, with more common bulbar-onset disease, cognitive impairment and FTD and a shorter survival [45^▪]. At the molecular level, nuclear TDP-43 depletion induces miss-splicing of UNC13A resulting in reduced UNC13A levels [46^▪▪], and this is favored by the at-risk allele in UNC13A[47^▪▪].

GENOTYPE-PHENOTYPE CORRELATIONS

Although it is not possible to discriminate sporadic ALS from specific subtypes of hereditary ALS on clinical grounds, several specific genotype-phenotype correlations exist (see Table 1). This suggest that certain genetic variants selectively affect the vulnerability of certain pools of neurons. Mutations in C9orf72 are more commonly associated with a bulbar onset ALS and with concomitant FTD and a more aggressive disease course. Mutations in SOD1 most frequently present a decade earlier than sALS, with a lower limb onset and predominant lower motor neuron involvement. The disease progression can be slow or fast, depending on the mutation [48^▪].

Mutations in TARDBP present more commonly with upper limb onset and typically have a younger disease onset but slow disease progression [49^▪]. Bulbar onset is uncommon for TARDBP-ALS. FUS mutations can present as aggressive juvenile ALS (in particular de-novo P525L mutations), or axial ALS with profound neck extensor weakness, or classic ALS with slower disease progression [50,51^▪]. Cognitive impairment is uncommon.

Mutations in TBK1 often present with FTD and patients presenting with ALS often develop frontotemporal involvement later in the disease [29,30].

GENE TESTING

With the advent of molecular therapies for specific genetic subtypes of ALS and ongoing gene therapy trials, there is a pressing need for better gene testing in ALS [52]. In patients with a family history of ALS, this has become common practice, but as many mutations in ALS genes have incomplete penetrance, testing apparently sporadic cases is important as well, irrespective of the age at disease onset [53]. There is a growing consensus to offer gene testing to all ALS patients, but the access to gene testing and the test modalities are variable geographically. Current guidelines state that all persons with ALS should be offered genetic testing, consisting of candidate gene testing for the most common genes, including C9orf72, SOD1, TARDBP and FUS at a minimum [54^▪▪]. Gene panel analysis following whole-exome or whole-genome testing is possible, but there is no consensus on which genes to include in such panels. Recent studies suggest that pathogenic or likely pathogenic variants can be detected in 10–25% of apparently sporadic cases [8^▪,10,55]. How to deal with new variants of unknown significance is a matter of debate, but the meaning of such variants will become clear with time as more and more results from gene panels become available. The turn-around time of a test is also an important factor to take into account to choose the method for gene testing and should preferably be less than 1–3 months, certainly for genes for which molecular therapies can be offered.

Gene testing should not be offered without access to genetic counseling and education. It should include a personalized risk assessment and discussion about the heritability of ALS and the potential impact of a positive test result preceding the decision to undergo gene testing [54^▪▪]. Posttest counseling should include discussions about the impact of negative, uncertain or positive test results, the risks for family members and the available options with regards to preventing transmission of gene mutations to future generations.

CONCLUSION

Our knowledge about the genetics of ALS has expanded significantly in the last two decades, but the genetic landscape appears to be complex. Pathogenetic mutations underly 15–20% of all ALS, but rare variants with smaller effect size, common variants with small effect size and combinations thereof contribute to the susceptibility of ALS. As the first genetic therapies are emerging, gene testing for all patients becomes an important issue, but the optimal testing approach remains less clear.

Acknowledgements

The authors thank our patients and their families for participating in our research studies.

Financial support and sponsorship

P.V.D. holds a senior clinical investigatorship of FWO-Vlaanderen (G077121N) and is supported by the E. von Behring Chair for Neuromuscular and Neurodegenerative Disorders, the ALS Liga België and the KU Leuven funds ‘Een Hart voor ALS’, ‘Laeversfonds voor ALS Onderzoek’ and the ‘Valéry Perrier Race against ALS Fund’.

Conflicts of interest

P.V.D. has served in advisory boards for Biogen, CSL Behring, Alexion Pharmaceuticals, Ferrer, QurAlis, Cytokinetics, Argenx, UCB, Muna Therapeutics, Alector, Augustine Therapeutics, VectorY, Zambon, Amylyx, Sapreme Technologies, Novartis (paid to institution). P.V.D. has received speaker fees from Biogen and Amylyx (paid to institution).

REFERENCES AND RECOMMENDED READING

Papers of particular interest, published within the annual period of review, have been highlighted as:

• ▪ of special interest

• ▪▪ of outstanding interest