Classifying the adverse effects of drugs can be done according to any number of criteria, but one approach we find useful is to subdivide the list of AEs into those in which adverse effects are therapeutically related or class-specific, and those that are more idiosyncratic or associated with a single therapeutic entity. This distinction is useful when charting a course of action to minimize an adverse effect. While there’s no convincing evidence that switching a patient from Lipitor to Zocor will alter risk of cataract, recommending a second-generation anti-histamine over a first-generation drug might be all that’s needed to alleviate a patient’s dry eye. In addition, AEs that are class-specific are more likely to be dose-dependent, so it may be possible to titrate medications to an optimal level where therapeutic effects are retained but untoward actions are eliminated or minimized.
Once a drug attains Food and Drug Administration approval and is brought to market, the process of identifying significant AEs begins in earnest. That may sound odd, but the drug approval process can only identify the most prominent of potential AEs, such as those that occur in short courses or in a high percentage of patients; these are usually modest for drugs that achieve final approval. With post-approval monitoring comes the ability to track use in millions of patients over years of use. Often, AEs that are identified following approval are then examined in controlled trials, either prospectively or using established longitudinal databases such as the Beaver Dam Eye Study.1 Availability of these databases allows investigators to address questions of long-term drug effects in large populations, and these questions often have surprising answers.
Another example of a therapeutic heavyweight with potential for ocular AEs is the bisphosphonates, drugs such as alendronate that are first-line therapy for prevention and treatment of osteoporosis. Ocular side effects associated with these drugs include anterior uveitis and scleritis, and although these effects do not seem to be related to the drugs’ mechanism of action (inhibition of bone resorption by osteoclasts) they are dose-related.5-7 In contrast to the example of statins, it would seem difficult to justify use of a drug for prevention of disease when it induces other, potentially more serious eye disorders. It’s important to note, however, that ocular AEs seem to be limited to the most potent of the bisphosphonates, particularly pamidronate and zoledronate.
An important class of drugs linked to ocular side effects is the thiazolidinediones, drugs used to treat type-2 diabetes that activate the peroxisome proliferator-activated receptor pathways involved in glucose utilization. A number of studies have linked these compounds to an increased risk of macular edema, but there is still some debate as to the significance of these effects.8,9 Since there are other drugs available, patients with other risk factors for macular disease might be best served by avoiding this class of diabetes medications.
The most common of all systemic drug side effects are, unquestionably, those referred to as anti-cholinergic. It might be more accurate to refer to these adverse responses as anti-muscarinic, since they result from the blockade of muscarinic cholinergic receptors of the parasympathetic nervous system.10 These include pathways that control heart rate, lacrimal and salivary secretion, urine flow and gastro-intestinal motility. Anti-cholinergic drug side effects are often the first place clinicians look when faced with a patient complaining of constipation, dry mouth or dry eye.
While anti-cholinergic AEs may be the most common, perhaps the most significant ocular AEs ophthalmologists deal with on a daily basis are those that stem from systemic glucocorticoid use.10 Ophthalmic use of steroids relies on topical formulations which allow for the combination of high doses and short courses that can mitigate the risks associated with these agents, including increases in intraocular pressure and cataract formation. In contrast, patients receiving long-term systemic steroid therapy need to be monitored for these serious actions of steroid use.
The Unexpected AE
Examples in recent literature show that, ultimately, it’s impossible to predict with certainty how each patient will respond to therapeutic intervention. Case reports of drugs with clearly defined mechanisms of action that elicit completely unpredictable AEs remind us that no therapeutic course of action is without risk. A recent report described an apparent drug-induced corneal ring infiltrate that progressed rapidly and responded to treatment efforts poorly.11 The patient was receiving an investigational anti-cancer drug (perifosine); in such cases it is not even possible to state with certainty that the drug was the causative agent. Many such case reports reflect extremely rare AEs, yet it’s worth remembering that the rare case can still be our patient.
A classic example of an idiosyncratic AE occurs with topiramate, a drug originally developed as an anti-epileptic that acts by interfering with voltage- and ligand-gated ion channels.12 While this mechanism of action is similar to many other drugs used for seizure disorders, only topiramate has been associated with cases of bilateral angle-closure glaucoma, which, although rare, is an ophthalmic emergency that can lead to loss of vision. Despite this, topiramate has gained a host of new indications in recent years, including migraine, bipolar disorder and neuropathic pain.
Another unexpected AE is the case of the anti-TNF-a mAb etanercept, a drug that is used in several types of inflammatory conditions (arthritis, psoriasis) yet has been associated with ocular inflammation, including uveitis and scleritis.13 These reactions occurred in patients with rheumatic disease but no sign of ocular involvement prior to etanercept therapy, and in all reported cases the condition resolved upon withdrawal of the drug.
Among the newer groups of biological therapeutics, epidermal growth factor receptor kinase inhibitors such as gefitinib, erlotinib, sorafenib and sunitinib have been associated with severe but rare cases of corneal perforation.14 These drugs are used to treat various solid tumors and represent a significant therapeutic advance over previous therapies. An interesting aspect of their mechanism of action stems from the targeting of tumor-specific genotyping of EGFR polymorphisms that may be useful in selecting which agent to use in specific patients;15 this same technique may hold promise as a means to predict those at risk for adverse effects, and thus provide a way to avoid the unfavorable sequelae of therapy.
The promise of personal genetics in medicine isn’t just about deriving ideal treatments to address an individual patient’s condition, but also to predict and to avoid potential drug AEs. This approach is already in development for mitigating risk of severe AEs such as Stevens-Johnson syndrome,16 but may be suitable for a more customized usage in the future. Genomic approaches, as well as the more traditional clinical rigor, are tools needed to uphold another of those medical clichés: Above all, do no harm.
Dr. Abelson is a clinical professor of ophthalmology at Harvard Medical School. Dr. McLaughlin is a medical writer at Ora Inc.
1. http://www.bdeyestudy.org/ accessed 2 Oct 2013.
2. Klein BE, Klein R, Lee KE, Grady LM. Statin use and incident nuclear cataract. JAMA 2006;295:23:2752-8.
3. Hermansa MB, Ahnb SA, Rousseau MF. Statin therapy and cataract in type 2 diabetes. Diabetes & Metabolism. 2011; 37:139–143.
4. Leuschen J, Mortensen EM, Frei CR, Mansi EA, Panday V, Mansi I. Association of statin use with cataracts: A propensity score-matched analysis. JAMA Ophthalmol 2013. doi: 10.1001/jamaophthalmol.2013.4575. [Epub ahead of print]
5. Pazianas M, Clark EM, Eiken PA, Brixen K, Abrahamsen B. Inflammatory eye reactions in patients treated with bisphosphonates and other osteoporosis medications: Cohort analysis using a national prescription database of systemic drugs with ocular adverse effects. J Bone Mineral Res 2013;28:455-463.
6. Dipika V. Patel DV, Horne A, House M, Reid IR, McGhee CN. The incidence of acute anterior uveitis after intravenous zoledronate. Ophthalmology 2013;120:773–776.
7. Etminan M, Forooghian F, Maberley D. Inflammatory ocular adverse events with the use of oral bisphosphonates: A retrospective cohort study. CMAJ 2012;184:8:E431-4.
8. Singh S, Segal JB. Thiazolidinediones and Macular Edema. Arch Intern Med 2012;172:1011-1013.
9. Idris I, Warren G, Donnelly R. Association between thiazolidinedione treatment and risk of macular edema among patients with type 2 diabetes. Arch Intern Med 2012;172:13:1005-1011.
10. Li J, Tripathi RC, Tripathi BJ. Drug-induced ocular disorders. Drug Safety 2008;31:127-141.
11. Keenan JD, Fram NR, McLeod SD, Strauss EC, Margolis TP. Perifosine-related rapidly progressive corneal ring infiltrate. Cornea 2010;29:583–585.
12. Abtahi MA, Abtahi SH, Fazel F, Roomizadeh P, et al. Topiramate and vision: A systematic review. Clin Ophthalmol 2012;6:117-31.
13. Gaujoux-Viala C, Giampietro C, Gaujoux T, et al. Scleritis: A paradoxical effect of etanercept? Etanercept-associated inflammatory eye disease. J Rheumatol 2012;39:2:233-9.
14. Borkar DS, Lacouture ME, Basti S. Spectrum of ocular toxicities from epidermal growth factor receptor inhibitors and their intermediate-term follow-up: A five-year review. Support Care Cancer 2013;21:4:1167-74.
15. Erdem L, Giovannetti E, Leon LG, Honeywell R, Peters GJ. Polymorphisms to predict outcome to the tyrosine kinase inhibitors gefitinib, erlotinib, sorafenib and sunitinib. Curr Top Med Chem 2012;12:15:1649-59.
16. Wei CY, Lee MT, Chen YT. Pharmacogenomics of adverse drug reactions: Implementing personalized medicine. Hum Mol Genet 2012;21:R1:R58-65.