Colorectal Cancer (CRC) is characterized by specific molecular and mutational alterations that play an important role in the choice of treatment. Approximately 40% of CRC patients have KRAS mutations, while about 6% have NRAS mutations leading to the constitutive activation of Ras-Raf-mitogen-activated protein kinase (MAPK) signaling pathway, downstream of the epidermal growth factor receptor (EGFR) (1, 2). As a result, many CRC patients are resistant to anti-EGFR therapies (3, 4).

Other relevant genomic alterations include mutations in the BRAF gene, which encodes a serine/threonine kinase within the MAPK signaling pathway. Approximately 8-12% of CRC patients have a BRAF mutation and 95% of BRAF mutations result in a substitution of the amino acid valine with glutamic acid at position 600 (BRAFV600E) (5–7). Patients with BRAF mutations generally have a poor prognosis in CRC due to their resistance to conventional therapies. Oral BRAF inhibitors (BRAFi) have had a significant impact on the treatment approach and clinical outcomes for patients with metastatic colorectal cancer (mCRC) who are not considered suitable candidates or have been unsuccessfully treated with standard therapies (8, 9).

Regorafenib (REG), an oral multikinase inhibitor, since it inhibits antigenic and oncogenic kinases, such as vascular endothelial growth factor receptors (VEGFR), platelet-derived growth factor receptors (PDGFR), fibroblast growth factor receptors (FGFR), and BRAF, received approval from the US Food and Drug Administration (FDA) in September 2012 for the treatment of mCRC (10, 11). In 2020, a pure oral BRAFi encorafenib (ENC) was approved in combination with cetuximab (CET) (12). Oral tyrosine kinase inhibitors (TKIs) offer a better prognosis in terms of progression free survival and overall survival than conventional therapies, with several advantages over injectable formulations, such as flexibility, convenience, cost-effectiveness, and better compliance (13, 14).

Despite the positive impact on patients’ survival in mCRC, the utilization of REG and ENC is not exempt from the occurrence of adverse events (AEs): 91% of patients treated with REG experienced AEs, including fatigue (46%), hand-foot skin reaction (HFSR) (42%), hypertension (30%), diarrhea (25%), and oral mucositis (25%) (15). Gastrointestinal disorders such as diarrhea (38%), nausea (38%), and decreased appetite (31%) were the most commonly reported AEs for ENC, along with fatigue (33%), dermatitis acneiform (30%), asthenia (24%), arthralgia (23%), and headache (20%). Moreover, serious AEs (SAEs) such as anemia (6%) and intestinal obstruction (5%) were primarily reported (5). The safety profile of oral TKIs is not fully understood, especially concerning neurological and psychiatric AEs and the consequent worsening of patients’ quality of life (QoL). A post-marketing study found an increased reporting of cerebral infarction, including ischemic stroke, for REG (16). A possible role in the pathophysiology of neurological disorders of oral kinase inhibitors cannot be ruled out. Other pharmacovigilance studies described a possible occurrence of peripheral neuropathies in patients under treatment with BRAFi and/or MEK inhibitors (MEKi) for melanoma (17, 18). However, no safety studies based on the spontaneous reporting system (SRS) databases aim to evaluate REG- and ENC-related neuropsychiatric ADRs in mCRC.

For all these reasons, the objective of this study was to assess and describe all ADRs associated with oral TKIs in the treatment of mCRC, focusing on neuropsychiatric ADRs, by conducting an analysis of the US Food and Drug Administration’s Adverse Event Reporting System (FAERS) database.

This study focused on the neuropsychiatric disorders associated with REG and ENC approved for the treatment of mCRC and reported in the FAERS database. The decision to evaluate this safety concern was motivated by the lack of SRS database studies that aimed to detect ADRs for TKIs in mCRC.

The distribution of reports over the years has been influenced by the timing of approval and clinical use of REG and ENC. Neuropsychiatric ADRs have predominantly been reported for REG. However, the recent increase in their reporting compared to all other reports, especially in 2021, may be attributed to the recent approval of ENC in combination with CET for mCRC (12). The frequency of neuropsychiatric ADRs with REG appeared to be higher than what was observed in a previous study that focused on neuropsychiatric ADRs with oral TKIs approved for gastrointestinal stromal tumor (GIST) and was based on the EudraVigilance database (28.6% vs. 13.0%) (16). This difference could be attributed to the higher incidence of CRC compared to GIST in the entire population, as well as the use of different SRS databases for the analysis. Looking at the characteristics of reports, neuropsychiatric ADRs were almost equally distributed by gender; however, they were mainly reported for adults than all other reports, especially for patients aged between 50 and 64 years. This could be explained by the rising trend of CRC diagnosis in young adults, which is facilitated by recommended screening methods enabling early detection of CRC (23). Nevertheless, a previous study reported that mental and cognitive disorders are more common in elderly patients diagnosed with CRC (24). The literature data on gender and neuropsychiatric ADRs are controversial. Some studies have reported that females with CRC are more prone to develop toxicity associated with targeted therapies than males (25, 26). However, serious neurological symptoms and mood disorders were mainly observed in men rather than women treated with target therapies (27). Additionally, neuropsychiatric ADRs were primarily reported by consumers, as stated in a previous study (28). Although FAERS is a global database, a large portion of spontaneous reports, including those related to neuropsychiatric ADRs, originated from America. This could potentially be attributed to the growing manufacturing activities and investments made by pharmaceutical companies leading to an increased request for efficient pharmacovigilance service (29).

Considering the seriousness and outcomes, a higher percentage of serious ADRs was shown for neuropsychiatric reports compared to all other reports, particularly those with disabled outcomes. In a pharmacovigilance study, serious neurological ADRs were mainly reported for ENC in combination with binimetinib in cases where melanoma was the indicated condition (17). Furthermore, REG seemed to have a higher likelihood of developing serious ADRs, including neuropsychiatric ones, in patients with mCRC as well as in subjects with GIST (16). This could potentially be explained by the invasive nature of cancer, but also by the lower survival benefit resulting from the oral TKI discontinuation after the occurrence of serious ADRs (5, 30).

The median TTO of neuropsychiatric ADRs was lower with REG compared to ENC. Based on the literature, the median TTO of ADRs related to the nervous system disorders was 225 days with ENC+CET (31) while it was 54 days for peripheral neuropathy with ENC+binimetinib (17). Another study investigating all neuropsychiatric ADRs associated with TKIs in GIST patients showed a median TTO of 22 days for REG (16). These findings were higher than it was observed in the present study, likely due to limited cases reporting TTO, the involvement of different types of cancer, or the presence of comorbidities.

Neuropsychiatric ADRs most commonly reported for REG included neuropathy peripheral, headache, dizziness, paraesthesia, and confusional state, as observed in several studies (10, 16, 32, 33). Headache was also frequently reported for ENC, with approximately 20% of patients receiving ENC+CET experiencing this ADR, as indicated in the literature (5, 31). Additionally, the FDA label for REG highlights headache as very common, without specifically mentioning of other neuropsychiatric ADRs. However, the European Medicines Agency (EMA) Summary of Product Characteristics (SmPC) for REG mentions tremor, peripheral neuropathy, and posterior encephalopathy syndrome (PRES) (34).

The disproportional analysis revealed some potential unexpected signals that were not reported in the FDA labels, with certain mechanisms yet to be defined. For REG, a statistically significant ROR was found for nerve disorders and associated symptoms, such as neuropathy peripheral, peripheral sensory neuropathy, sciatica, neuralgia, hyperaesthesia, thermohyperaesthesia, hypoaesthesia, and burning sensation. Although REG is related to low direct neurotoxicity, it could have potential side effects targeting VEGF, including vascular toxicity that could lead to neurotoxicity as a consequence (35). The onset of nerve disorders may occur as a result of unexpected stimulation of the MAPK pathway, leading to increased growth of Schwann cells. Conversely, selective activation of the MAPK signaling pathway or, alternatively, overexpression of RAF showed a negative impact on Schwann cell differentiation (36, 37). Moreover, aberrant immune activation against peripheral nerves could lead to treatment-induced inflammatory demyelinating peripheral neuropathy, particularly due the significant role played by the MAPK pathway in the production of proinflammatory cytokines (38, 39). However, a potential correlation with pre-existing nervous or psychiatric comorbidities, as well as the concomitant use of other drugs cannot be excluded. Also, prior treatment lines, including cytotoxic drugs that may cause neurotoxicity, such as oxaliplatin, should be taken into account when reporting neuropsychiatric ADRs in pretreated CRC patients (40). Furthermore, the role of prior radiation therapy and the administration of immune checkpoint inhibitors (ICIs) could impair neuronal repair and exacerbate nerve toxicity with REG (41, 42). Moreover, hyperaesthesia, thermohyperaesthesia, hypoaesthesia, and burning sensation, can be considered as consequential symptoms of neuropathy, as already known in the literature (18).

A disproportionality analysis revealed a significant association between REG and taste disorders, including hypogeusia. Disturbed taste is a well-known ADR of REG and mentioned in the EMA SmPC but not in the FDA Prescribing Information. In an exploratory study, taste disorders were reported in 55% of patients treated with REG and were found to be associated with a lower QoL (43). The onset of taste disorders, such as hypogeusia, may be linked to an uncharacterized neuropathic etiology, possibly attributed to the involvement of the chorda tympani nerve responsible for taste and salivary innervation (44). Another significant and unexpected disproportion was observed in relation to slow speech. Speech disorders may be associated with the development of stomatitis, which is caused by DNA damage to the mucosal surface (45); otherwise, a case report described a patient undergoing treatment with REG who experienced slurred speech, probably due to the presence of brain metastases (46). Focusing on motor dysfunctions, such as dysstasia, several case reports (46–48) have suggested a potential cause: the interference of TKIs with cerebral signal pathway or distress resulting from their antiangiogenic action, which may lead to brain-related complications in patients with predisposing conditions. Moreover, a genetic polymorphism in pharmacokinetic and pharmacodynamic pathways could also contribute to these ADRs (46).

Regarding sleep disturbances, there was a significant unknown disproportionality for poor-quality sleep, hypersomnia and, insomnia within the SOC psychiatric disorder, for REG. TKIs are commonly associated with sleep disorders, as evidenced by a clinical study where 14% of TKI-treated patients had serious sleep disturbances, with improvements noted upon TKI discontinuation (49). Furthermore, the onset of sleep disturbance is frequently reported as a troublesome symptom in patients undergoing REG treatment (50). However, the onset of sleep disorders may be related to the pathogenesis and progression of CRC itself; in fact, more than 70% of CRC patients have reported sleeping issues associated with circadian disruption. The dysregulation of circadian genes in cancer leads to the downregulation of PER2 and the upregulation of β-catenin protein levels, which can cause the proliferation of CRC cells (51). Additionally, the development of insomnia may not be specific to the type of drug used; a previous study found no differences among chemotherapy, ICIs, and target therapy (52–54).

The consistent disproportionality of altered state of consciousness, incoherent status, disorientation, and delirium is not reported in the FDA label for REG. Similarly, cognitive disorders are not mentioned in the FDA label for ENC. Nevertheless, altered mental status is reported in the EMA SmPC for REG (34). Cognitive impairment can be considered as an effect resulting from the activation of multiple signaling pathways, including RAF/MAPK and VEGF/VEGFR2/mTOR, which subsequently affect various cellular processes associated with corticogenesis (33, 41, 55). Cognitive disorders have been also found to be three to five times higher in CRC patients compared to healthy control, with higher rates of impairment observed in women than men (56).

Moreover, depressed mood was identified as a potential unknown disproportionality signal for both REG and ENC. Depressed mood may be associated with a diagnosis of depression following the diagnosis of CRC, whether or not it is related to the administration of BRAFi (57, 58). The onset of depressed mood with ENC may be linked to its concomitant use with CET, as mentioned in CET FDA label (59). Depression could also be influenced by the impact on QoL of typical CET toxicities, for instance diarrhea and skin rash. It’s worth noting that CET-related skin rash occurs much less frequently and is less severe when combined with ENC compared to monotherapy (60). However, a possible correlation with ENC cannot be excluded.

Considering brain oedema, it is an unknown effect associated with ENC that could be also caused by brain metastases rather than ENC. Localized or peripheral oedema was reported in the EMA SmPC (61), but not in the FDA label. Oedema was frequently observed with BRAFi/MEKi combinations (62), but no literature evidence was found for the association with CET.

This study confirms the crucial role of the FAERS database in evaluating neuropsychiatric ADRs related to oral TKIs approved for mCRC. REG has higher probability of reporting neuropsychiatric ADRs compared to ENC, which could be attributed to its previous approval for mCRC. The case/non-case analysis and the calculation of ROR have highlighted some ADRs that have not been extensively reported in the literature but are worth discussing, such as nerve disorders, taste disorders, slow speech, and sleep disturbances for REG, and cognitive disorders, brain oedema, and depressed mood for ENC.

Neuropsychiatric ADRs can significantly impact a patient’s QoL and treatment outcomes. By considering pre-existing nervous or psychiatric comorbidities, oncologists can better assess the potential risks and take appropriate measures to minimize the occurrence of ADRs or manage them effectively. However, further real-world studies are necessary to achieve a better understanding of these ADRs and their impact on patients’well-being.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://fis.fda.gov/extensions/FPD-QDE-FAERS/FPD-QDE-FAERS.html.

The study was approved by the Ethics Committee of University Hospital of Messina (n. 40/23 of 28th February 2023). Written informed consent from the participants was not required to participate in this study in accordance with the national legislation and the institutional requirements.

MAB: Conceptualization, Formal Analysis, Methodology, Visualization, Writing – original draft, Writing – review & editing. GR: Conceptualization, Formal Analysis, Methodology, Visualization, Writing – original draft. EES: Formal Analysis, Visualization, Writing – original draft. GC: Visualization, Writing – original draft. TF: Visualization, Writing – review & editing. MS: Visualization, Writing – review & editing. DS: Visualization, Writing – review & editing. ES: Supervision, Validation, Visualization, Writing – review & editing. NS: Supervision, Validation, Visualization, Writing – review & editing.