Migraine, a chronic neurological disorder, causes severe headaches and can potentially lead to disability. Symptoms like nausea and sensitivity to light and sound can last up to 72 h (Ashina, 2020). Migraine is associated with psychiatric disorders, atopic diseases, and cardiovascular disease, among others, with mutual influences among these conditions (Terhart et al., 2025). Migraine affects over one billion people worldwide (GBD, 2016 Headache Collaborators, 2018), predominantly adults (Leonardi et al., 2020). The high disability rate associated with migraine significantly impairs patients’ quality of life (QoL) and productivity, creating significant financial strain on individuals and the community (GBD, 2016 Headache Collaborators, 2018; Ashina et al., 2021).

The 2018 International Classification of Headache Disorders, Third Edition (ICHD-III) (Arnold, 2018) provides standardized diagnostic criteria, thereby enhancing the accuracy of migraine diagnosis and aiding in its ongoing management and treatment. Around 40% of individuals with episodic migraine may benefit from preventive therapy to reduce migraine frequency and delay progression, yet only 17% receive it (Silberstein et al., 2025; L et al., 2007). Migraine treatment is categorized into acute and preventive treatments, with a preventive treatment assessment recommended after each acute episode (Kumar and Kadian, 2025). Medications for migraine prophylaxis mainly include anticonvulsants, β-blockers, angiotensin receptor blockers/ACE inhibitors, calcium antagonists, serotonin antagonists, antidepressants, CGRP monoclonal antibodies (mAbs), onabotulinumtoxin A, and gepants (Puledda et al., 2024a). Additionally, coenzyme Q10 (CoQ10), riboflavin, omega-3 fatty acids, melatonin, and vitamin D3 have shown potential in migraine prevention (Tseng et al., 2024; Puliappadamb et al., 2022; Sazali et al., 2021; Okoli et al., 2019). Agents such as memantine may also exert beneficial effects through modulation of glutamate signaling (Charles, 2021). CGRP drugs, typically administered by monthly or quarterly injection, may improve patient adherence due to their lower dosing frequency (Varnado et al., 2022; Sacco et al., 2022). However, injections can be inconvenient, requiring professional administration and potentially causing injection site reactions (Mechtler et al., 2022; Suzuki et al., 2025). Moreover, these drugs are unavailable in most countries globally (Kiarashi et al., 2021; Tana et al., 2024; Puledda et al., 2024b), highlighting the unique advantages of oral preventive treatments.

Although prophylactic treatments for migraine have demonstrated effectiveness and safety, most evidence focuses on individual drugs. There is a lack of direct comparisons between different medications (Versijpt et al., 2024; Raffaelli et al., 2023; Deligianni et al., 2023). This limitation in evidence makes it difficult for clinical guidelines to provide a clear hierarchical structure for the selection of prophylactic medications (Arnold, 2018). Specifically, most existing studies fail to directly compare the efficacy and safety of different prophylactic drugs, which poses challenges for clinicians in selecting appropriate treatments. Therefore, more research is needed to fill this knowledge gap and offer more specific guidance for clinical practice.

Since adults are primarily affected by migraine (Leonardi et al., 2020), this study evaluated the efficacy and safety of oral prophylactic drugs in migraine patients aged 18–65 years using a network meta-analysis (NMA). The upper age limit was set at 65 because patients aged 65 and above are generally considered elderly, a group that typically has multiple comorbidities and different drug metabolism and tolerability compared with younger patients (Jhaveri et al., 2014; Ngcobo, 2025). Therefore, to ensure the accuracy and comparability of the study results, this study focused on adults aged 18–65. Future studies should consider a broader age range to better address the needs of different patients.

This network meta-analysis aimed to compare the efficacy, safety, and tolerability of oral preventive drugs in adults aged 18–65 years with migraine, providing evidence-based guidance for clinical treatment selection.

This research report, including randomized clinical trials (RCTs), was registered in the International Prospective Register of Systematic Reviews (PROSPERO, CRD42024621316, 13 December 2024) and adhered to the principles of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher et al., 2009; Liberati et al., 2009; Page et al., 2021).

Among 17,443 citations retrieved, the titles and abstracts of 11,350 citations were screened after removing duplicates. After screening the full texts of 376 articles, the systematic review and NMA included 44 clinical trial articles (4,612 participants) (Carrieri et al., 1988; Steiner et al., 1988; Stewart et al., 1988; Tuca et al., 1989; Gawel et al., 1992; Peikert et al., 1996; Schoenen et al., 1999; Lucking et al., 1988; Bussone et al., 1999; Diener et al., 2001a; Diener et al., 2001b; Pradalier et al., 2001; Edwards et al., 2003; Silvestrini et al., 2003; Mei et al., 2004; Sandor et al., 2005; Silberstein et al., 2006; Ashtari et al., 2008; Afshari et al., 2011; Bostani et al., 2013; Sadeghi et al., 2015; Assarzadegan et al., 2016; Noruzzadeh et al., 2016; Choudhary et al., 2017; Hesami et al., 2018a; Hesami et al., 2018b; Gazerani et al., 2019; Hajihashemi et al., 2019; Shanmugam et al., 2019; Ghorbani et al., 2020; Chowdhury et al., 2022; Hedayat et al., 2022; Sherafat et al., 2022; Balali et al., 2024; Mehramiri et al., 2024; Mosarrezaii et al., 2025; Roghani et al., 2024; Diener et al., 1996; Keskinbora and Aydinli, 2008; Luo et al., 2012; Gonçalves et al., 2016; Askari et al., 2017; Ebrahimi-Monfared et al., 2017; Khani et al., 2021) (Supplementary Figure S1). Among these, 37 studies (Carrieri et al., 1988; Steiner et al., 1988; Stewart et al., 1988; Tuca et al., 1989; Gawel et al., 1992; Peikert et al., 1996; Schoenen et al., 1999; Lucking et al., 1988; Bussone et al., 1999; Diener et al., 2001a; Diener et al., 2001b; Pradalier et al., 2001; Edwards et al., 2003; Silvestrini et al., 2003; Mei et al., 2004; Sandor et al., 2005; Silberstein et al., 2006; Ashtari et al., 2008; Afshari et al., 2011; Bostani et al., 2013; Sadeghi et al., 2015; Assarzadegan et al., 2016; Noruzzadeh et al., 2016; Choudhary et al., 2017; Hesami et al., 2018a; Hesami et al., 2018b; Gazerani et al., 2019; Hajihashemi et al., 2019; Shanmugam et al., 2019; Ghorbani et al., 2020; Chowdhury et al., 2022; Hedayat et al., 2022; Sherafat et al., 2022; Balali et al., 2024; Mehramiri et al., 2024; Mosarrezaii et al., 2025; Roghani et al., 2024) and 7 studies (Diener et al., 1996; Keskinbora and Aydinli, 2008; Luo et al., 2012; Gonçalves et al., 2016; Askari et al., 2017; Ebrahimi-Monfared et al., 2017; Khani et al., 2021) were two-arm and three-arm RCTs, respectively. The features of the included studies are detailed in Supplementary Table S1 18 studies (1,602 patients) (Ashtari et al., 2008; Afshari et al., 2011; Bostani et al., 2013; Sadeghi et al., 2015; Assarzadegan et al., 2016; Noruzzadeh et al., 2016; Hesami et al., 2018a; Hajihashemi et al., 2019; Ghorbani et al., 2020; Hedayat et al., 2022; Sherafat et al., 2022; Balali et al., 2024; Mehramiri et al., 2024; Mosarrezaii et al., 2025; Roghani et al., 2024; Askari et al., 2017; Ebrahimi-Monfared et al., 2017; Khani et al., 2021) were conducted in Iran, constituting approximately 34.7% of the total participants. The bias risk assessment indicated that, among the included studies, 59% were rated as having a medium risk of bias, and 41% were rated as having a low risk of bias (Supplementary Table S1). Supplementary Figure S2, S3 illustrate the factors influencing the overall risk of bias score (Mcguinness and Higgins, 2021).

In this NMA, we evaluated oral medications for migraine prevention. Topiramate, propranolol, and valproate significantly reduced migraine frequency, intensity, and duration, consistent with the 2019 European Headache Federation guidelines, which classify topiramate and propranolol as Level A recommendations and valproate as Level B (Sacco et al., 2019). These agents are also recognized as first-line options in the 2024 International Headache Society Global Practice Recommendations, which synthesize existing evidence on established oral therapies alongside emerging biologics (Puledda et al., 2024a). Despite their efficacy, these medications are associated with higher rates of adverse events, highlighting the need for careful risk–benefit assessment in clinical practice. This study supports the potential role of memantine, melatonin, and vitamin D3 in the prevention of migraine. Although these drugs are not traditionally considered preventive medications, their multi-target mechanisms offer a broader range of therapeutic options for specific patient populations. Specifically, memantine may alleviate migraine attacks by modulating the glutamate signaling pathway, thereby reducing neuronal hyperexcitability (Charles, 2021). Melatonin may indirectly reduce the frequency and pain intensity by regulating circadian rhythms and improving sleep quality (Ong et al., 2018; Nagtegaal et al., 1998). Vitamin D3 may demonstrate significant efficacy in improving migraine by modulating neurotransmitter levels and reducing inflammatory responses (Tseng et al., 2024). However, it is important to note that evidence for these agents is primarily derived from small or single studies, which limits the generalizability of these findings. Therefore, future studies should further explore the mechanisms of action and clinical efficacy of these drugs to verify their long-term effects and safety in migraine prevention.

In addition, this study systematically evaluated the efficacy and safety of various combination therapies. The results showed that flunarizine plus topiramate, valproate plus magnesium, and folic plus pyridoxine were significantly more effective in reducing the frequency, intensity, and duration of migraine attacks compared to monotherapy, and were associated with favorable safety profiles. These combination therapies may generate synergistic effects through distinct mechanisms of action, thus enhancing their preventive efficacy. For example, the combination of valproate and magnesium may enhance the antimigraine effects of valproate by supplementing intraneuronal magnesium ion levels (Khani et al., 2021). Due to the limitations of current network meta-analysis models, the specific contribution of each drug in combination therapies cannot be clearly distinguished. While valproate, topiramate, and amitriptyline are well-established first-line treatments, the added value of adjuncts like magnesium remains uncertain. However, some of these combinations were supported by limited evidence, and larger trials are needed to confirm these findings. Future studies should further investigate the long-term effects and potential side effects of these combination therapy regimens, particularly their efficacy and safety across different patient populations. Investigating the long-term impact of these combination therapies on patients’ QoL, as well as determining the optimal dosage and treatment duration, will be important directions for future research.

In the analysis of the 50% response rate, moderate heterogeneity was observed (I² = 72.7%), and a random-effects model was used to account for between-study variability. Excluding studies with less than 16 weeks of follow-up (Silberstein et al., 2006; Afshari et al., 2011; Bostani et al., 2013; Choudhary et al., 2017) reduced heterogeneity (I² = 50.7%). In contrast, low heterogeneity (I² < 25%) was found for migraine frequency, duration, pain intensity, and quality of life (QoL), supporting the use of a fixed-effect model for these outcomes. The consistency across studies for these endpoints suggests reliable results. The heterogeneity in the 50% response rate may be influenced by follow-up duration; therefore, future studies should standardize follow-up periods to improve comparability. This study found that topiramate, flunarizine, propranolol, valproate, amitriptyline, cinnarizine, and nortriptyline were associated with improved QoL. However, given that the evidence is derived from a limited number of studies, these findings should be interpreted cautiously in clinical practice. Although drugs such as valsartan and a-dihydroergocryptine have shown therapeutic effects, these findings are largely based on individual studies and require further investigation to be confirmed.

34.7% of the participants were from Iran, which may introduce regional bias and limit the generalizability of our findings. Although the included Iranian studies were methodologically sound and met our inclusion criteria, differences in demographics, genetics, or healthcare practices compared to other populations may affect the external validity. Future studies should include more diverse populations to confirm the applicability of these findings across different regions and clinical contexts.

Rimegepant and Atogepant have been approved for migraine prevention (Tepper, , 2020; Deeks, 2022). However, our systematic literature review found no RCTs focusing on these drugs for migraine prevention in patients aged 18–65 years. Some studies have covered a broader age range, such as 18–80 years (Schwedt et al., 2022). This suggests that these drugs may have significant clinical value for a wider patient population. Focusing on patients aged 18–65 years may limit the applicability of our findings. Future research should include a broader age range, such as individuals aged 18 and above, for a more comprehensive perspective. It should also focus on age-stratified studies to better understand the drugs’ therapeutic effects and tolerability.

This study not only validated the efficacy of traditional drugs such as topiramate, valproate, and propranolol, but also focused on uncovering the potential value of non-traditional preventive medications and combination therapies. These agents and combination strategies offer novel approaches to migraine prevention, especially for patients who cannot tolerate conventional therapies. Future studies should further investigate the mechanisms of action and clinical efficacy of these agents to validate their long-term effectiveness and safety in migraine prophylaxis. In addition, given the multi-target nature of these drugs, exploring differences in their efficacy across various patient subgroups will be of considerable importance. QoL measures should be more comprehensively incorporated into future studies to better evaluate the impact of treatment on patients’ daily functioning and overall wellbeing.

Although this study comprehensively evaluated the effects of various oral medications on migraine prevention in adults, it has several limitations. First, variations in the initial measurements of certain endpoints (e.g., frequency, intensity, MIDAS score, and duration) may have affected the reliability of the findings. Second, the validity of the NMA was limited by the inclusion of some medications tested in fewer than 100 patients, which may have introduced small-study bias and influenced the reported results. Third, the evaluation of outcomes such as QoL was based on a restricted number of studies, leading to insufficient statistical power. Fourth, Iran contributed the largest proportion of participants (approximately 34.7% of the total 4,612), which may represent an additional potential source of regional bias. Fifth, this study did not include newer oral preventive agents such as rimegepant and atogepant, which have recently been approved for migraine prevention. This was primarily due to the limited number of RCTs in the predefined age group (18–65 years). These drugs represent emerging treatment options and should be considered in future comparative studies with broader inclusion criteria. In addition, some medications included in this analysis, such as indobufen and cinnarizine, are not approved by the US Food and Drug Administration (FDA), although they have been commonly prescribed in several countries outside the US.

This network meta-analysis confirms the significant efficacy of traditional agents such as topiramate, valproate, and propranolol in reducing migraine frequency, pain intensity, and duration. However, these medications are associated with a higher incidence of adverse events, highlighting the need for careful risk–benefit assessments. Additionally, the study found that non-traditional medications such as memantine, melatonin, and vitamin D3 may have potential therapeutic effects. Combination therapies, such as flunarizine plus topiramate and valproate plus magnesium, demonstrate superior efficacy and a lower incidence of adverse events compared to monotherapy. When traditional treatments are not tolerated or prove ineffective, combination therapies or non-traditional medications may be considered. For high-risk patients, treatment decisions should prioritize drug safety. Future research should further evaluate the long-term effectiveness and safety of non-traditional preventive medications and combination therapies, particularly in different patient subgroups and in relation to QoL. Overall, this study provides important support for clinical decision-making and highlights potential directions for future research on migraine prevention.