Table 1 Main characteristics of the studies included in the analysis. General characteristics of studies included in the analysis. OR = odds ratio; CI = confidence interval; P = P value; HWE = Hardy–Weinberg equilibrium.

Lemos et al., 2010 [42]

Case-control study. The Authors state that sample size is a limitation of the study

not providing enough power to detect a variant with an

OR < 1.5.

Diagnosis according to the International Headache

Society (IHS), using the International Classification of Headache Disorders

(ICHD) I (before 2004) and II. N = 188

People with no history of migraine, age-matched to cases N = 287

Portuguese

Genomic DNA extraction from peripheral blood

leucocytes

rs1553005 (CGRP)

1) Investigate the role

of BDNF in migraine susceptibility; 2)

CGRP, studied for the first time as candidate gene in migraine; 3)

possible interaction of BDNF and CGRP genes in migraine’s susceptibility.

GG: OR (95 %CI) 1.00

0), p= -; multivariable logistic regression analysis

GC: OR (95 %CI )1.40 (0.94–2.08), p = 0.09; multivariable logistic regression analysis

CC: OR (95 %CI) 0.86 (0.42–1.77), p = 0.69; multivariable logistic regression analysis

Case and control groups in HWE for the polymorphism

rs1553005(CGRP)*rs2049046(BDNF) = GC*AT – OR 1.88 (95 %CI 1.20–2.93), P = 0.005

Menon et al., 2011

[43]

Case-control study. Power analysis reported

Clinical diagnosis by an

experienced clinical neurologist based on international criteria N = 278

Individuals

with no family history of migraine, age, sex and ethnicity matched to cases N = 322

Australian (east coast) Caucasian:of

European descent living in Australia,

with ancestors emigrated within the last 160 years from

the British Isles and other parts of Europe

Genomic DNA extraction from white blood cells

rs35815751 (16 bp deletion in intron 1 of the CALC A gene)

Invetigate the role of 16 bp deletion in the first

intron of the CALCA gene in the risk of migraine

II (homozygous, insertion/insertion); no deletion/deletion OR (95 %CI) 1.2 (0.74–1.85) in migraine, for genotypes (P = 0.575) and for alleles

(P = 0.502); logistic regression analysis

ID (heterozygous, insertion/deletion); no deletion/deletion OR (95 %CI) 1.1 (0.68–1.79) in migraine with aura, (genotypes, P = 0.666; alleles, P = 0.7); logistic regression analysis

DD (homozygous,

deletion/deletion);

no deletion/deletion OR (95 %CI) 1.6 (0.70–3.44) in migraine without aura, (genotypes, P = 0.325;

alleles, P = 0.276); logistic regression analysis

Genotypes in HWE (migraineurs: P_HWE=0.73 and controls: P_HWE=0.247)

No association between rs35815751 and migraine for genotypes (P = 0.575) nor alleles

(P = 0.502), and migraine with aura (genotypes, P = 0.666; alleles, P = 0.7) or without aura (genotypes, P = 0.325;

alleles, P = 0.276)

Guldiken et al., 2013 [44]

Case-control study on females.

Diagnosis of migraine by neurologist based on the

criteria of ICHD-II. N = 134

Healty volunteers, health care personal and

postpartum females who were hospitalized in the obstetric clinic. N = 96

Local (Turkish)

DNA isolation from peripheral blood

rs3781719 (SNP T-692 C of CALC A gene)

Investigated the

frequency of CALCA T-692 C

in migraineurs and its association to migraine attack

frequency and severity

TT: 42.5 %

TC: 42.5 %

CC: 14.9 %

HWE has been evaluated.

No

association between the genotype and allele

frequency of migraine (P = 0.44), without and with aura (P = 0.52), and the severity and frequency of migraine attacks. Frequency of migraine attacks has been measured as the number

of attacks in a month. The severity has been assessed

with the visual analog scale

Sutherland et al., 2013

[45]

Case-control study. Power analysis reported

Diagnosis by a clinical neurologist according to the IHS criteria. N = 284

Matched for sex, age

(+/−5 years) and ethnicity. N = 284

South

Eastern Australian: adult Caucasians of

European descent living in Australia, with ancestors emigrated

within the last 160 years from British

Isles and other parts of Europe

DNA extraction from white

blood cells

rs3781719 (SNP 624 (T/C) of the CALC A gene promoter)

Investigate the contribution of the 4 selected SNPs in the risk of migraine

TT:

-migraine47.8 %;

-migraine with aura 49.0 %

-migraine without aura 45.7 %

TC: -migraine44.7 %;

-migraine with aura 42.2 %;

-migraine without aura 49.4 %

CC: -migraine7.5 %; -migraine with aura 8.8 %;

-migraine without aura 4.9 %

HWE has been verified. Genotypes for SNPs were in HWE in case and

control groups

No significant association with migraine (P = 0.260), migraine with aura (0.563) and migraine without aura (0.133)

rs145837941 (4218T > C

base-exchange in the coding sequence of

CALC A)

TT: 94.9 %

TC: 4.7 %

CC: 0.4 %

No significant association with migraine (P = 0.913)

rs3754701 (SNP in the RAMP1 gene promoter at position −

1166 (T/A))

TT: - migraine 37.4 %; - migraine with aura 42.1 %; - migraine without aura 30.6 %

TA: - migraine 49.4 %; - migraine with aura 44.8 %; - migraine without aura 56.1 %

AA: - migraine 13.2 %; - migraine with aura 13.1 %; - migraine without aura 13.3 %

No significant association with migraine (P = 0.360), migraine with aura (P = 0.276) and migraine without aura (P = 0.260)

rs7590387 ((G/C) 1.4 kb downstream of the

RAMP1 gene)

GG: - migraine 25.7 %; - migraine with aura 23.9 %; - migraine without aura 28.4 %

GC: - migraine 54.4 %; - migraine with aura 54.3 %; - migraine without aura 54.5 %

CC: - migraine 19.9 %; - migraine with aura 21.7 %; - migraine without aura 17.0 %

No significant association with migraine (P = 0.341), migraine with aura (P = 0.566) and migraine without aura (P = 0.299)

Cargnin et al., 2015 [46]

Case-control study. Power calculation reported

Diagnosis fulfilling

ICHD-II criteria for migraine without aura

A

for at least 1 year and for medication overuse headache (MOH) or previous or current

diagnosis of MOH.Patients with migraine without aura and patients with MOH. N (migraine without aura) = 219; N (MOH) = 130

Matched

by age and sex and

of same

ethnicity. N = 209

Italian (north-west)

DNA extraction from

peripheral blood

rs3781719 [CALC A (T > C)]

Investigate the role of the two selected SNPs in: 1) risk

of inconsistent response to triptans; 2) risk of

trasformation of episodic migraine to MOH

TT: (A) OR (95 %CI) 0.96 (0.62–1.48), P = 0.84; logistic regression analysis adjusted for triptan (frovatriptan vs. other triptans), and for

polymorphisms, associated with triptan response. A = log-additive model of inheritance; D = dominant model

of inheritance; R = recessive model of inheritance

TC: (D) OR (95 %CI) 1.04 (0.59–1.85), P = 0.88; logistic regression analysis adjusted for triptan (frovatriptan vs. other triptans), and for

polymorphisms, associated with triptan response. A = log-additive model of inheritance; D = dominant model

of inheritance; R = recessive model of inheritance

CC: (R) OR (95 %CI) 0.80 (0.30–2.11), P = 0.65; logistic regression analysis adjusted for triptan (frovatriptan vs. other triptans), and for

polymorphisms, associated with triptan response. A = log-additive model of inheritance; D = dominant model

of inheritance; R = recessive model of inheritance

HWE tested. P_HWE rs3781719 = 0.52; P_HWE

rs3754701 = 0.77; P_HWE rs7590387 = 0.41. For only MOH patients:

(P_HWE rs3781719 = 0.18; P_HWE rs3754701 = 1; P_HWE

rs7590387 = 0.69)

No significant association for endpoint 1 [risk

of inconsistent response to triptans (being consistent responders defined as patients experiencing

a ≥ 2 point reduction

after triptan administration in at least 2 out of 3 consecutive

attacks)]; significant association of RAMP1rs7590387GG [OR (95 %CI) (R) 0.27 (0.13–0.57) P = 0.0002)] for endpoint 2 (risk of

trasformation of episodic migraine to MOH)

rs3754701 [RAMP 1 (T > A)]

TT; (A) OR (95 %CI) 0.90 (0.60–1.37), P = 0.63; logistic regression analysis adjusted for triptan (frovatriptan vs. other triptans), and for

polymorphisms, associated with triptan response. A = log-additive model of inheritance; D = dominant model

of inheritance; R = recessive model of inheritance

TA; (D) OR (95 %CI) 0.73 (0.41–1.32), P = 0.30; logistic regression analysis adjusted for triptan (frovatriptan vs. other triptans), and for

polymorphisms, associated with triptan response. A = log-additive model of inheritance; D = dominant model

of inheritance; R = recessive model of inheritance

AA; (R) OR (95 %CI) 1.19 (0.53–2.67), P = 0.66; logistic regression analysis adjusted for triptan (frovatriptan vs. other triptans), and for

polymorphisms, associated with triptan response. A = log-additive model of inheritance; D = dominant model

of inheritance; R = recessive model of inheritance

rs7590387 [RAMP 1 (C > G)]

CC: (A) OR (95 %CI) 1.11 (0.75–1.65), P = 0.60; logistic regression analysis adjusted for triptan (frovatriptan vs. other triptans), and for

polymorphisms, associated with triptan response. A = log-additive model of inheritance; D = dominant model

of inheritance; R = recessive model of inheritance

CG: (D) OR (95 %CI) 0.84 (0.46–1.54), P = 0.57; logistic regression analysis adjusted for triptan (frovatriptan vs. other triptans), and for

polymorphisms, associated with triptan response. A = log-additive model of inheritance; D = dominant model

of inheritance; R = recessive model of inheritance

GG: (R) OR (95 %CI) 1.69 (0.86–3.30), P = 0.12; logistic regression analysis adjusted for triptan (frovatriptan vs. other triptans), and for

polymorphisms, associated with triptan response. A = log-additive model of inheritance; D = dominant model

of inheritance; R = recessive model of inheritance

Wan et al., 2015 [48]

Exploratory analysis. Limitations reported by the Authors: small sample, the analysis could not be conducted on the specific

tissue derived from trigeminovascular or cerebral biopsies,the blood DNA was from multiple cell lineages with likely different DNA methylation

Migraineurs recruited at International

Headache Center of Chinese PLA General Hospital. N = 26

Matched healthy controls with no significant

differences in gender and age compared to migraineurs

group. N = 25

Patients recruited at International

Headache Center of Chinese PLA General Hospital

DNA extraction from EDTA blood samples obtained through

the cubital vein

Analysis of CpG islands of the RAMP1 gene in a 3000 bp region including putative

promoter sequences and the first exon

Investigate if the methylation

pattern of the promoter of RAMP1 gene in

peripheral leukocyte is associated with migraine

Low methylation trend without statistical significance in migraine vs. control (total average methylation level: 8.41 % ±1.92 % vs. 9.90 % ± 3.88 %, P = 0.197)

The ROC the curve and Youden criterion have been used

for the determination of the optimum cut-off of methylation

level

No significant difference in the DNA

methylation pattern of RAMP 1

between migraine and

control groups

Moreno-Mayordomo et al., 2019

[47]

Prospective, observational, multicentre, study enrolling female participants. Sample power calculation reported

Diagnosis of chronic migraine according to ICHD- III edition, beta version. classified as responders to onabotulinum toxinA (defined as defined as a reduction of at least 50 %

in the number of monthly migraine days 3 months after the second administration. N = 117

Diagnosis of chronic migraine according to ICHD- III edition, beta version. classified as non responders to onabotulinum toxinA. N = 33

Caucasian ethnicity and Spanish origin

Genomic DNA extraction from peripheral blood anticoagulated in EDTA-K₃

rs3781719 (CALC A)

Investigate the effect of rs3781719 on

the response to OnabotulinumtoxinA

TT: 53.85 % responders and

27.27 % non responders

TC:

38.46 % responders and

63.63 % non responders

CC:

7.69 % responders and

9.09 % non responders

All of the variants wer in HWE

Apart from rs222749 of TRPV 1 gene.

Significant differences in risk to lack of response to onabotulim toxin A for the SNP rs3781719 of gene

CALC A: OR (95 %CI) (dominant model) 3,11 (1,33 − 7,26), (codominant model) 1,6 (0,85 − 3,0), (recessive model) 1,2 (0,31 − 4,71)