Injectable disease-modifying therapies can cause a wide range of injection site reactions, from transient erythema and eczemalike reactions to severe ulcers and necrosis, which may require treatment cessation and surgical intervention. The most common injection site reactions include erythema, bruising, induration, immune-mediated inflammatory reactions, lipoatrophy, cutaneous necrosis, and ulceration.⁹ Rare cutaneous reactions have been reported, including psoriasis exacerbation,¹³ granulomatous dermatitis with focal sarcoidal features,¹⁴ Raynaud phenomenon,¹⁵ panniculitis,¹⁶,¹⁷ vasculitis,¹⁸ subacute cutaneous lupus erythematosus,¹⁹ morphea,²⁰ localized pigmentation disorders,²¹ cutaneous mucinoses,²² and fixed drug eruptions.²³ Injection site reactions most commonly occur during the first month of treatment but have been reported up to 29 months after initiating therapy.²⁴ Some reports suggest a decrease in the incidence with increasing treatment duration.²⁵ Injection-site reactions are observed more frequently in injection sites with reduced subcutaneous fat (i.e., thighs and arms) than in areas with increased subcutaneous fat (i.e., buttocks and abdomen).²⁶ There is a higher predominance in females than in males, ranging from 2:1 to 8:1, although this may be an artifact of the higher overall incidence of MS in women.²⁴,²⁷

Along with flulike symptoms, injection site reactions are the most common side effect observed in patients using injectable IFN-β, occurring in up to 90% of patients using subcutaneous formulations and up to 33% of those using intramuscular formulations.⁹,²⁸ In three controlled clinical trials investigating the use of IFN-β 1b (Betaseron®), injection site reactions occurred in 86% of patients compared with 37% with placebo. Injection site necrosis occurred in 5% of patients compared with none
in the placebo cohort. Other reactions, including inflammation (53%), pain (18%), hypersensitivity (3%), and edema (3%) were significantly associated with IFN-β 1b treatment. Approximately 76% of patients who developed injection site reactions did so in the first 3 months of treatment.²⁹ In addition, a placebo-controlled randomized trial of IFN-β 1b for secondary progressive MS reported injection site reactions in 43.6% of patients versus 10.3% with placebo. Necrosis occurred in 4.7% compared with zero in the placebo group and inflammation in 50% compared with 4.2% in the placebo group.³⁰ A 30-month postlicensure study reviewed the adverse event reports for IFN-β 1b following Food and Drug Administration (FDA) approval in 1993. Adverse events included erythema (51%), pain (30%), and necrosis (13%), with 6% of patients with injection site reactions subsequently discontinuing therapy and 21% of patients with injection site necrosis requiring surgical intervention.²⁴

In two multicenter studies evaluating the safety and efficacy of IFN-β 1a (Rebif®) in patients with relapsing remitting MS, injection site reactions occurred in 89% and 92% of patients taking 22 µg and 44 µg three times weekly, respectively, compared with 39% with placebo. Injection site necrosis occurred in six patients (3%) taking 44 µg three times weekly and two patients (1%) taking 22 µg three times weekly over the course of 2 years.³¹

Intramuscular administration (Avonex®) is associated with fewer skin reactions when compared with subcutaneous administration of IFN-β. In pooled placebo-controlled clinical studies, injection site pain, inflammation, and injection site reactions occurred in 8%, 6%, and 3% of patients, respectively, compared with 6%, 2%, and 1% with placebo. To date, there have been no reports of necrosis with the use of Avonex®. However, lipoatrophy can occur when improperly injecting intramuscular formulations subcutaneously, due to an incorrect injection angle.⁹,²⁸

Panniculitis, the inflammation of subcutaneous adipose tissue, has been described in patients treated with IFN-β therapy, with subsequent lipoatrophy seen in up to 46% of patients.⁹,³² Localized lipoatrophy is characterized by loss of subcutaneous adipose tissue in the area of the injection site, resulting in well-circumscribed areas of skin depression, seen most commonly on the anterolateral surface of thighs or upper arms. Histopathology shows atrophic and diminutive fat lobules in the subcutaneous fat (Figure 9.1). Proposed mechanisms include a direct toxic effect on adipocytes, inciting an inflammatory response followed by a hypersensitivity reaction and residual loss of subcutaneous fat.³³ Patients generally present with pain, erythema, and induration at injection sites of the arms or thighs (abdomen and buttocks are less common). A minority of reactions
are severe enough to cause difficulty with ambulation or require surgical debridement.³⁴ Reports of both septal and lobular panniculitis have been described with or without accompanying vascular thrombosis.¹³,¹⁶,¹⁷,³²,³⁵,³⁶,³⁷,³⁸,³⁹,⁴⁰ Lipoatrophy is a concerning side effect, as it may be irreversible, resulting in disfiguring contours of the thighs and arms. It is more commonly observed with glatiramer acetate injection than with IFN-β. There are multiple potential explanations for the occurrence of lipoatrophy with IFN-β, including incorrect injection into the dermis, secondary reaction from the high immunogenicity of IFN-β, and administration of nonprewarmed medication.¹⁰,²²

A rare and more recently reported cutaneous side effect of IFN-β therapy is localized scleroderma, or morphea. Morphea is a cutaneous fibrosing connective tissue disorder characterized by excessive collagen deposition leading to thickening of the dermis, subcutaneous tissue, or both (Figure 9.2). Cases of morphea have been reported with both IFN-β 1a and IFN-β 1b.²⁰,⁴¹,⁴² In one case, woody induration consistent with morphea appeared on the anterior thighs 6 months after injection of IFN-β 1a.²⁰ Another case reported a morpheaform reaction to IFN-β 1b injection on the thighs and abdomen after 10 years of IFNβ-1b therapy.⁴² Histopathologically, these cases were consistent with morphea, with thickened collagen bundles extending into the subcutaneous fat, paucity of adnexal structures, and a deep reticular lymphocytic infiltrate (Figure 9.3).⁴¹ The authors proposed a possible dysregulation of inflammatory cytokines, trauma, or a combination of both as the underlying cause.

Injection site ulceration and necrosis typically occur within the first 4 months of therapy but have been reported greater than 1 year after initiating therapy. Necrosis is generally limited to an area 3 cm or less in diameter with extension into the subcutaneous fat, although larger lesions with involvement of the fascia overlying muscle have been reported.²⁹ Healing is usually associated with scarring. In rare cases, debridement with subsequent skin grafting may be required. Histologically, perivascular and interstitial lymphohistiocytic infiltrates, panniculitis, and thrombosis of deeper vessels have been reported.¹³,²⁹,⁴³,⁴⁴,⁴⁵,⁴⁶,⁴⁷,⁴⁸,⁴⁹,⁵⁰ Proposed mechanisms include thrombosis and necrosis of dermal vessels secondary to abnormal aggregation of platelets as well as a hypersensitivity reaction within blood vessels.⁴³,⁵¹,⁵²,⁵³ Incorrect injection technique and improper depth of injection have been identified as risk factors.⁵⁴ The decision of whether to continue treatment depends on the extent of necrosis. Some patients have experienced healing of necrotic skin lesions while continuing therapy. Patients should be advised not to administer therapy into the affected area.²³ If multiple lesions are present, the general consensus is that therapy should be discontinued.²⁹

The pathogenesis of injection site reactions with IFN-β therapy is not fully understood. Direct toxic and proinflammatory mechanisms, as well as IgE-mediated and delayed hypersensitivity reactions have been proposed as potential causes. Furthermore, immune-mediated necrotizing vasculitis and platelet-dependent thrombosis may play a role. Patients with MS have clotting abnormalities at baseline secondary to abnormal platelet activation, which may be further exacerbated by the immunomodulatory treatments used in MS.³⁷,⁵¹,⁵⁵ IFN-β administered subcutaneously can induce inflammatory skin reactions due to local chemokine induction with immune cell extravasation. More specifically, studies have demonstrated an upregulation of chemokines (CCL2 and CXCL10), with recruitment of circulating T cells to evolving skin lesions.⁵⁶ Histologically, reports have demonstrated a lupuslike reaction with superficial and deep perivascular and periadnexal lymphocytic infiltrate with focal interface damage and interstitial mucin.²³,⁵⁷,⁵⁸ In severe cases, thrombosis of deeper vessels may be seen.¹³,²⁹,⁴³,⁴⁴,⁴⁵,⁴⁶,⁴⁷,⁴⁸,⁴⁹

Rare systemic reactions have been reported with the use of IFN-β therapy in patients with MS, although a direct correlation between disease onset and the use of IFN therapy could not be proven in all cases. Twelve cases
of sclerosing skin disorders, including both limited and diffuse cutaneous sclerosis, have been reported in patients with MS.⁵⁹ In five of these patients, IFN-β had been administered between 1 and 8 years before the onset of sclerosis, which the authors suggested may have been the trigger.⁵⁹ One case of cutaneous vasculitis with renal impairment was reported in a patient injecting IFN-β for 10 weeks who developed purpura, proteinuria, and hematuria after an injection.¹⁸ In another case, a middle-aged female developed disseminated cutaneous lesions after her third IFN-β injection, which reappeared after a subsequent injection and resolved upon discontinuation of therapy.⁶⁰ A diffuse maculopapular rash was reported in one patient after her second IFN-β injection, which resolved with antihistamines and corticosteroids but reappeared after the third injection.⁶¹ In addition, there was one report of severe, new-onset dermatomyositis in a male patient after 5 years of IFN-β therapy. There was a temporal association between exacerbation of symptoms and repeated injections, with in vitro studies revealing enhanced type I IFN signaling in lymphoid cells in response to IFN-β.⁶² Furthermore, IFN-induced sarcoidosis is a rare reaction but has been reported in seven patients with MS, as well as one with multiple myeloma and one patient with renal cell carcinoma.¹⁴,⁶³,⁶⁴,⁶⁵,⁶⁶,⁶⁷,⁶⁸,⁶⁹,⁷⁰ In one case, a patient developed noncaseating granulomas in her skin and in pulmonary lymph nodes after 3 years of therapy. The authors proposed that the development of sarcoidosis was due to a dysregulation in the modulatory role of IFN-β (and more generally type I interferon) expression in chronic inflammation.⁶⁴ In addition, multiple reports of new-onset psoriasis, psoriatic arthritis, and psoriasis exacerbations have been reported with the use of IFN therapy, with some lesions lasting as long as 6 years after the cessation of IFN-β.¹³,⁷¹,⁷²,⁷³,⁷⁴ The mechanism is thought to be due to IFN-β’s proinflammatory effect in psoriasis, with upregulation of the interleukin (IL)-23/T helper cell (Th)17 pathway.⁷²,⁷⁵

Injection site reactions have been reported in 20% to 80% of patients taking glatiramer acetate, including erythema (66%), inflammation (49%), pain (73%), and pruritus (40%).⁷⁶,⁷⁷,⁷⁸ These injection site reactions are generally mild or moderate in severity and disappear spontaneously in hours to days. Similar rates of injection site reactions are seen with the 20- and 40-mg dosing regimens.⁷⁶ Aviv et al recently proposed three distinct histologic responses to glatiramer acetate injections: (1) erythematous plaques and nodules with superficial and sometimes deep inflammatory infiltrates seen in an acute response, (2) morpheaform plaques with perivascular and interstitial dermatitis and thick collagen bundles seen in chronic cases, and (3) lipoatrophy initially present as deep infiltrates of lymphocytes, neutrophils, eosinophils, and plasma cells, leading to a more chronic infiltration,
loss of subcutaneous adipose tissue, and delicate fibrosis.⁷⁹ Average onset of injection site reactions after glatiramer acetate therapy is 1.75 years, but reactions may occur as early as the first injection.⁷⁹ Possible causes of injection site reactions include the high immunogenicity of glatiramer acetate, repeated trauma to the fat inciting an inflammatory reaction, as well as an allergic reaction to glatiramer acetate components.⁸⁰,⁸¹

Glatiramer acetate is characteristically associated with frank panniculitis and subsequent lipoatrophy.³³,⁸²,⁸³,⁸⁴,⁸⁵,⁸⁶ Of all the injectable disease-modifying therapies for MS, glatiramer acetate is associated with the highest prevalence of lipoatrophy. This occurrence was originally reported to be rare, occurring in 2% of patients in clinical studies, but in a recent study involving full skin examinations in patients taking glatiramer acetate, prevalence was as high as 45%.⁸⁴,⁸⁷ Trauma alone can cause lipoatrophy, such as the lipoatrophy seen with acupuncture. It is thought that this type of lipoatrophy is caused by the activation of macrophages and release of cytokines such as tumor necrosis factor (TNF) and IL-1.⁸² Histologically, acupuncture-induced lipoatrophy is characterized by an involutional pattern with thin and elongated fat lobules, with only a few surrounding macrophages. However, on histopathologic examination of glatiramer acetate-associated lipoatrophy, a lobar panniculitis predominates with histiocytes engulfing lipids from necrotic adipocytes and T lymphocytes in the fat lobules. Thickened septa with scattered lymphoid follicles composed of B lymphocytes may also be seen.³³ This type of adverse event appears to be independent of injection technique.⁸⁴ Glatiramer acetate panniculitis is thought to be secondary to the drug’s high immunogenicity, as a robust immune response is required for glatiramer acetate to be therapeutic in MS treatment.³³,⁸⁰,⁸⁵ In mild cases, cutaneous lesions may resolve with treatment discontinuation but can reappear if injections are restarted. However, two recent reports suggested that glatiramer acetate-associated lipoatrophy can persist and even progress despite discontinuing treatment.⁸⁰,⁸⁸