Critical Reviews in / 66 (2008) 155–162

Clinical significance and treatment of skin rash from erlotinib in non-small cell lung cancer patients: Results of an Experts Panel Meeting C. Gridelli a,∗, P. Maione a, D. Amoroso b, M. Baldari c, A. Bearz d, V. Bettoli e, E. Cammilluzzi f, L. Crino` g, F. De Marinis h, F.A. Di Pietro i, F. Grossi i, D. Innocenzi j, G. Micali k, F.V. Piantedosi l, M. Scartozzi m a Division of Medical Oncology, “S.G. Moscati” Hospital, Contrada Amoretta, 83100 Avellino, Italy b Division of Medical Oncology, Versilia Hospital, Lido di Camaiore, Italy c Division of , University of Genova, Italy d Division of Medical Oncology A, CRO, Aviano, Italy e Division of Dermatology, Arcispedale Sant’Anna, University of Ferrara, Italy f Division of Oncology, Sandro Pertini Hospital, Roma, Italy g Department of Medical Oncology, Perugia Hospital, Perugia, Italy h V Division of Oncological Pneumology, S.C. Forlanini Hospital, Roma, Italy i Division of Dermatology, Marchesi Hospital, Indago, Milano, Italy j Department of Dermatology and Plastic , La Sapienza University, Roma, Italy k Division of Dermatology, University of Catania, Italy l Day Hospital Oncology, V. Monaldi Hospital, Naples, Italy m Clinic Oncology, Ospedali Riuniti, Ancona, Italy Accepted 18 October 2007

Contents

1. Introduction ...... 156 2. Skin rash from erlotinib and other HER1/EGFR targeted agents ...... 156 2.1. Definition and pathological/clinical findings ...... 156 2.2. Clinical significance and relationship with response and survival ...... 157 2.3. Other dermatologic side-effects of EGFR inhibitors...... 158 3. Skin rash management ...... 159 4. Suggestions by the Experts Panel ...... 159 4.1. Treatment ...... 159 4.2. Dose reduction and treatment interruption ...... 160 5. Suggestions for clinical research ...... 160 6. Conclusions ...... 160 Reviewers ...... 161 Conflicts of interest...... 161 References ...... 161 Biography ...... 162

Abstract Advances in the knowledge of tumor biology and mechanisms of oncogenesis has granted the singling out of several molecular targets for non-small cell lung cancer (NSCLC) treatment. Among these targets, epidermal growth factor receptor (EGFR), or HER1, has received particular attention in lung cancer treatment. Erlotinib, an orally available inhibitor of EGFR tyrosine kinase in a phase III randomized placebo-controlled trial (BR.21), has been proven to prolong survival in NSCLC patients after first or second line chemotherapy. Skin rash is

∗ Corresponding author. Tel.: +39 0825 203574; fax: +39 0825 203556. E-mail address: [email protected] (C. Gridelli).

1040-8428/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.critrevonc.2007.10.004 156 C. Gridelli et al. / Critical Reviews in Oncology/Hematology 66 (2008) 155–162 the most common adverse event associated with erlotinib treatment and it is often cause of negative impact on patients’ quality of life. There is no specific treatment for this toxicity due to the lack of evidence-based data and recommendations. A panel of Italian oncologists, who had participated to clinical trials and to the Expanded Access Program for erlotinib in NSCLC treatment, and dermatologists with experience with cutaneous toxicity from EGFR inhibitors, attended a Meeting held in Rome on December 2006 to discuss skin rash from erlotinib and to provide suggestions for managing this frequent side-effect. © 2007 Elsevier Ireland Ltd. All rights reserved.

Keywords: Erlotinib; Skin rash; NSCLC; EGFR; EGFR inhibitors

1. Introduction is a common side-effect of all HER1/EGFR inhibitors [9]. Rash affecting the skin above the waist, is the most com- Lung cancer is the leading cause of cancer-related mor- mon adverse event associated with erlotinib, and generally tality in both men and women [1], with 1.2 million new develops within 7–10 days of starting treatment [7]. Skin rash cases diagnosed every year and with 1 million deaths being may spontaneously resolve and reappear and it is reversible recorded worldwide in 2001 [2]. Non-small cell lung cancer following drug discontinuation. However, when it develops, (NSCLC), accounts for approximately 80% of all lung can- this chronic side-effect is very distressing for the patient. The cers. The majority of NSCLC patients present with advanced increasing use of EGFR-targeted agents and specifically of disease at diagnosis and a large part of those diagnosed with erlotinib in NSCLC treatment, and simultaneously the lack early stage disease eventually recur, experiencing metastatic of clinical trials on this topic, makes rash management and disease. For advanced disease palliation and the patients’ etiology investigation high priorities. quality of life are still the primary goal of , with A panel of Italian oncologists, who had participated to total cure remaining elusive. Advances in the knowledge of clinical trials and to the Expanded Access Program for tumor biology and mechanisms of oncogenesis has granted erlotinib in NSCLC treatment, and dermatologists with the singling out of several molecular targets for NSCLC treat- experience with cutaneous toxicity from EGFR inhibitors, ment. Among these targets, epidermal growth factor receptor attended a Meeting held in Rome on December 2006 to dis- (EGFR), or HER1, has received particular attention in lung cuss skin rash from erlotinib and to provide suggestions for cancer treatment. EGFR is a transmembrane receptor found managing this frequent side-effect. primarily on cells of epithelial origin. Autophosphorilation This article reports on the suggestions from this meeting of its intracellular domain initiates a cascade of events lead- and gives a brief summary of published information. How- ing to cell proliferation. EGFR is commonly expressed at a ever, because of the lack of trials evaluating rash , high level in a variety of solid tumors and it has been impli- evidence-based treatment recommendations are not yet pos- cated in the control of cell survival, proliferation, metastasis sible. The suggestions are only intended to assist the future and angiogenesis [3]. The main pharmacological strategies in investigation of agents to manage rash and to help clinicians clinical development for therapeutic inhibition of EGFR are when making decisions regarding patient management. monoclonal antibodies to antagonize ligand-receptor bind- ing, and small-molecules to inhibit tyrosine kinase domain activation [4]. The main EGFR inhibitors are cetuximab, an 2. Skin rash from erlotinib and other HER1/EGFR anti-EGFR monoclonal antibody, and erlotinib an EGFR- targeted agents tyrosine kinase inhibitor. The key indications for clinical use are colorectal cancer and head and neck cancer for cetuximab 2.1. Definition and pathological/clinical findings and NSCLC for erlotinib [5–7]. Erlotinib, a quinazolinamine, is a highly potent, orally available, reversible inhibitor of A variety of terms are used to describe HER1/EGFR EGFR tyrosine kinase. Erlotinib, in a phase III randomized inhibitors rash as , acneiform skin reaction, acneiform placebo-controlled trial, has been proven to prolong sur- rash, acneiform follicular rash, acne-like rash, maculopapu- vival (6.7 months versus 4.7 months for erlotinib and for lar skin rash, monomorphic pustular lesions [10]. However, placebo respectively, p = 0.001) in NSCLC patients after first rash does not have similar histology to acne and should not or second line chemotherapy [7]. The analysis of quality be described as acne, acneiform, or acne-like, because it of life and time to deterioration of patients reported symp- is predominantly pustular, not associated with comedones, toms showed statistically and clinically meaningful benefit and sebaceous glands are not affected. Therefore, use of for patients randomized to erlotinib [8]. Moreover, erlotinib terms “acne, acne-like, or acneiform” are discouraged, and resulted active (response rate of 8.9%) and safe (only 5% of rash should be described in phenotipic terms as “pustu- patients discontinued treatment for toxicity). Following this lar/papular rash, pustular eruption or follicular /intrafollicular trial, erlotinib has been approved by Food and Drug Admin- pustular eruptions” [11]. Recently, authors from the Derma- istration and Committee for Medicinal Products for Human tology Service of Memorial Sloan-Kettering Cancer Center use in chemotherapy-pretreated advanced NSCLC. Skin rash have described key clinical features of common dermatologic C. Gridelli et al. / Critical Reviews in Oncology/Hematology 66 (2008) 155–162 157 reactions to EGFR inhibitors, focusing mainly on the skin Table 1 toxicity and have discussed the and possible causes Skin rash/desquamation categories in NCI-CTC version 3 [10]. Grade 1 Macular or papular eruption or erythema without The more commonly affected areas are the face (nose, associated symptoms cheeks, nasolabial folds, chin and forehead) [9], the upper Grade 2 Macular or papular eruption or erythema with pruritus or other associated symptoms; localized desquamation or chest and/or back and, less frequently on the scalp, arms other lesions covering <50% of the body surface area and/or legs, abdomen and buttocks, but sparing palms, soles Grade 3 Severe, generalized erythroderma or macular, papular, or and mucosa. There may be caudal progression of the lesions. vescicular eruption; desquamation covering >50% of the From histopathologic studies of the papular pustular body surface area rash, two major pattern of reaction are identified: moderate Grade 4 Generalized exfoliative, ulcerative, or bullous dermatitis Grade 5 Death superficial dermal inflammatory cell infiltrate surrounding hyperkeratotic and ectatic follicular infundibula, particularly in the upper portion of the hair follicle, and superficial neu- covering <50% of the body surface area; grade 3 as severe, trophilic suppurative , with associated rupture of generalized erythroderma or macular, papular, or vescicular the epithelial lining. There are no apparent changes in dermal eruption, with desquamation covering >50% of the body sur- capillaries, eccrine glands or sebaceous glands. Microbiolog- face area; grade 4 as generalized exfoliative, ulcerative, or ical stains and cultures have failed to show an infectious cause bullous dermatitis. [12,13]. These findings suggest that the suppurative inflam- For what concerns the incidence of erlotinib-related skin mation may be a reaction to follicular rupture, or a direct rash, data from 485 patients enrolled onto the phase III effect of EGFR inhibition on normal follicular keratinization. randomized trial (BR.21) of erlotinib versus placebo in pre- Furthermore, these histopatologic findings demonstrate that treated advanced NSCLC show that skin rash occurred in the “acne-like” rash associated with EGFR inhibitors is not 75% of the patient population [7,15]. The median time to acne vulgaris, which is characterized by androgen-activated onset was 8 days (range: 1–113 days). Patients who had sebaceous gland hypertrophy and increased sebum produc- dose reductions because of rash were 10%; patients who tion, comedone development, and immunologic reactivity to had dose interruptions lasting more than seven consecutive colonization by Propionibacterium acnes. Despite the usually days because of rash were 7%, and more than 14 consecutive sterile nature of the pustules in these reactions, secondary days 3%. Rash was the cause of definitive discontinuation of bacterial infection with Staphylococcus aureus may super- erlotinib-therapy in 1% of the patients. The severity of most vene, especially in patients with nasal carriage and extensive rash conditions was mild to moderate. Sixty-six percent of mid-face involvement. the patients in the erlotinib arm experienced a worst severity The inhibition of EGFR in skin leads to a succession of of grade 1 or 2. Grades 3 and 4 rash occurred in 8% and <1%, events that eventually involves many cell types and results respectively, of erlotinib-treated patients (Table 2). in clinical manifestations that can be staged chronologically In other series, the incidence of dose reductions and inter- [14]. The initial manifestations consist of oedema, redness ruptions has been described as higher, and this discrepancy and a burning sensation on the face and upper trunk. These may be attributed to the lack of experience and skills of some symptoms are consistent with initial inhibition of EGFR in oncologists to manage this frequent side-effect. basal keratinocytes, which leads to the release of chemokines, causing the recruitment of leukocytes, vascular dilation and 2.2. Clinical significance and relationship with response increased permeability, all of which cause the activation of and survival nociceptive fibres leading to the sensory disturbance. Then occurs the papulopustular stage. The histological correlates The skin reaction is increasingly considered an indirect of papules and pustules consist of increased numbers of marker of relevant in vivo EGFR targeting. Pharmacokinetic inflammatory cells (T lymphocytes followed by abundant neutrophils) and debris in the superficial dermis, within and Table 2 around follicles, accompanied by varying degrees of oedema Skin rash in BR.21 trial and vasodilation. Finally we have the crusting stage. Res- Erlotinib (N = 485 patients) olution of often tender papulopustules is accompanied by Incidence of skin rash (any grade) 75% the drying of purulent material and its components, which Incidence of grade 1–2 skin rash 66% include neutrophilic and keratinocytic debris, parakeratotic Incidence of grade 3 skin rash 8% cells, fibrin and serum, and the formation of crusts. Incidence of grade 4 skin rash <1% Median time to onset (days) 8 (range 1–113) The National Cancer Institute Common Toxicity Criteria Patients with dose reductions because 10% (NCI-CTC) version n. 3 categorization of skin rash (Table 1) of rash defines grade 1 rash as macular or papular eruption or ery- Patients with dose interruption lasting 7% thema without associated symptoms; grade 2 as macular or more than 7 days because of rash papular eruption or erythema with pruritus or other associ- Patients with dose interruption lasting 3% more than 14 days because of rash ated symptoms, with localized desquamation or other lesions 158 C. Gridelli et al. / Critical Reviews in Oncology/Hematology 66 (2008) 155–162 studies have shown that rash parallels saturation of recep- is a superficial inflammation of the epithelium tor clearance and may be used as an indirect measure of of the lateral nail wall. This periungual inflammation, with effective receptor saturation and blockade at the higher dose crusted lesions along nail folds, occurs in 12–16% of patients levels or as an indicator for optimum biologic dose [16]. treated with EGFR inhibitors, after 4–8 weeks of therapy However, this raises the question of why about one third [25,26]. A high incidence of paronychia both with cetux- of patients do not develop skin toxicity, even at the max- imab and erlotinib therapy has been recently described in imal tolerated dose, which is likely to achieve effective a prospective study [27]. It is often described as a painful, EGFR saturation. It has been suggested that the variabil- erythematous inflammation at the lateral wall of the nail, ity in skin toxicity may be related to pharmacokinetic and usually at digit 1 of hands or feet. This can progress into pharmacodynamic differences or pharmacogenetic hetero- a lateral paronychia-like reaction with granulation tissue genity and EGFR polymorphisms among patients [17–19]. formation, which is very painful and presents as ingrown An alternative explanation is that the reaction may also be an nails. Secondary impetiginisation, painful fissures and pyo- indicator of a patient’s ability to mount an immune-mediated genic granulomas can occur [28]. These nail disorders are inflammatory response to the drug, thus becoming a sur- highly resistant to therapy and paronychia often needs several rogate marker of immunocompetence rather than receptor months to heal after withdrawal of treatment. Symptomatic inhibition. In addition, cyclo-oxygenase-2 (COX2) expres- relief has been reported with AlCl3 galenic gel [10]. Pre- sion within tumor tissue has also been associated with severity ventive measures are mandatory and they involve wearing of rash [20]. of wide shoes, adequate nail manicure and pedicure and Another interesting topic related to skin rash from EGFR hygienic advice to prevent secondary impetiginisation. Local inhibitors and specifically erlotinib, is a possible positive cor- antiseptics and antibiotic can prevent this impetiginisation. relation between rash and response and/or survival [21,22]. When it dose occur, oral antibiotics should be advised, If skin rash was found to be an important clinical surrogate although these antibiotics will not have an effect on the marker of antitumor activity or therapeutic efficacy, the reac- paronychia as such. Extraction of the nail or coagulation of tion would serve to identify the subgroup of patients more the granulation tissue can be taken into consideration ion likely to benefit from erlotinib. Such a tool to predict response cases of very painful and therapy resistant nail disorders [28]. to erlotinib may be useful also to identify the optimal biologic It is assumed that paronychia may result from the effect of dose of the drug, in order to maximize the clinical benefit. In EGFR inhibition on nail matrix keratinocytes, but the patho- fact clinical trials are ongoing on dose escalation of erlotinib genesis is unknown. until occurrence of grade 2 or maximally tolerated rash, Several weeks after start of EGFR inhibitory treatment respectively, to determine clinical benefit at the maximally in about 35% of patients, a dry itchy skin (xerosis cutis) on tolerated dose. In three phase II trials of erlotinib in patients the arms and legs arises [28]. When this condition aggra- with advanced NSCLC, advanced head and neck squamous vates, an asteatotic eczema, eventually complicated with cell carcinoma and advanced ovarian cancer, a significant a secondary impetiginisation by S. aureus or Herpes sim- correlation between rash and survival was found, with the plex infection, can occur. If the dry skin is manifested at median duration of survival increasing with severity of rash the hands or feet, it can lead to painful finger and toe-tips [23]. The time taken to develop rash was similar in all three (pulpitis sicca) and to fissures at the dorsal sides of the studies (the median time of rash onset was day 10, 8, and 7, in interphalangeal joints. Patient’s education with preventive the aforementioned three tumor types, respectively). In addi- measurements prior to beginning therapy is mandatory. Cen- tion, when the data from these three trials were pooled, there tral in the advice is avoidance of soaps, limiting shower was a trend between higher grades of rash and response, and time, the usage of luke-warm water and the frequent use a significant correlation between rash severity and survival of emollients [28]. Dry skin can be attributed to abnormal [23]. However, at the moment, the major retrospective evi- keratinocyte differentiation leading to a disturbed stratum dences on the correlation between occurrence of skin rash corneum and interference of sebaceous gland function, which and increased survival have been found for cetuximab in translates into a loss of the water-retaining function of the colon cancer treatment [24]. Further prospective evidences epidermis [14]. are needed to clarify the relationship between skin rash and EGFR signalling is involved in the normal hair cycle [29]. clinical benefit in NSCLC patients treated with erlotinib. Regulatory abnormalities in hair growth that are due to EGFR inhibitors are dependent upon the type and location of the hair, with alopecia and decreased growth on the scalp and 2.3. Other dermatologic side-effects of EGFR inhibitors extremities but increased growth (), distortion and thickness of hair on the face and eyelashes (trichomegaly) EGFR inhibitors, both anti-EGFR monoclonal antibodies [14]. Whereas alopecia has been reported after several months and tyrosine kinase inhibitors as erlotinib, are associated with with the use of EGFRIs as single agents (in 5% of patients), other skin- and skin-adnexal toxicities. These side-effects are the incidence of abnormalities at other sites has not been nail disorders (paronychia), xerosis cutis, hair and mucosal systematically recorded [14]. No adequate therapy for hair disorders. disorders is available at the moment. C. Gridelli et al. / Critical Reviews in Oncology/Hematology 66 (2008) 155–162 159

Mild to moderate mucositis (oral and nasal aphthous ideally, each stage of the skin rash should be managed with ulcers, dry anal or genital mucosa) have been reported with separate and specific approaches based on biological patho- several EGFR inhibitors [10]. However, only symptomatic genesis, beginning from the first stage of redness, oedema treatment with oral gels, or nasal and vaginal creams is and sensory disturbance. Lacouture suggests to use topi- possible. cal anaesthetics (as pramoxine and lidocaine) in the first stage of sensory disturbance. In the papulopustular stage, he considers a treatment with the topical anti-inflammatory 3. Skin rash management cyclosporine analogues pimecrolimus and tacrolimus or with topical [14]. Advantages of these calcineurin No controlled clinical trials of agents to manage rash inhibitors (pimecrolimus and tacrolimus) include the absence have been done. Thus, there is no specific treatment for the of the side-effects that are common with topical corticos- rash caused by EGFR agents. Only anecdotal evidence exists teroids. Moreover, Lacouture also suggests to consider in the and no aetiology-based and evidence-based treatment rec- papulopustular stage, if necessary, oral corticosteroids and ommendations have been published. When intervention is tetracyclines. Among tetracyclines, in Lacouture’s opinion, indicated, patients have been treated empirically with vary- doxycycline may be the best candidate because in addition ing response by drying agents, topical antiseptics, topical to its class-specific anti-inflammatory properties, it has anti- antibiotics (clindamycin and fusidic acid) systemic antibi- angiogenic properties and induces apoptosis in colorectal otics (tetracycline, minocycline, erythromycin and fusidic cancer cells [14]. Another approach proposed by Galimont- acid), topical retinoids (tretinoin), topical immunomodula- Collen et al. is NCI-CTC grade-specific [28]. These authors tory agents (pimecrolimus) and short-term topical steroids suggest only local therapy with metronidazol for grade 1 (hydrocortisone, triamcinolone acetonide and betametha- skin rash. For grades 2 and 3 they suggest local therapy sone dipropionate) or systemic steroids for the more with metronidazol plus systemic therapy with tetracyclines intense reactions [10,30–33]. In January 2004 a Forum on (minocycline or doxycycline). Only for grade 3 they add HER1/EGFR-inhibitor rash management was held in New a local therapy with wet wraps (NaCl 0.9% solution). For York City [11]. These authors, although suggesting that top- grades 2 and 3, in case of infection they suggest to use flu- ical corticosteroids are largely ineffective in patients with cloxacillin and in case of itch antihistamines. For grade 4 skin severe rash, have not excluded that they may have some effi- rash they suggest to refer to a burn-wound unit. cacy if used early in patients wiht mild rash. In this Forum the Preliminary data from a trial using 0.025% alpha- use of retinoids was not advised as their skin-drying effects hydroxy acid cream with 0.166% gentamicin and 0.122% are likely to exacerbate rash. On the contrary, studying topical betamethasone cream for grade 2 lesions showed improve- immunomodulatory agents (such as pimecrolimus [Elidel®]) ment, but the effectiveness of this combination treatment was recommended considering the inflammatory nature of remains in question as the trial was uncontrolled [34]. the rash. Antihistamines were considered by the Forum to pal- Roe` et al. described, prospectively, and managed, cutaneous liate a disturbing and chronic symptom like pruritus, although side-effects in 30 patients during cetuximab and erlotinib their role is marginal. Finally, these authors suggested to treat treatments [27]. Treatment of skin rash included topical treat- secondarily infected rash with a short course of oral antibi- ment with antiseptic soaps, antibiotics (topical erythromicin otics, preferring tetracyclines such as minocycline because or clindamycin), class II corticosteroids (prednicarbate) or a of their proposed weak anti-inflammatory effects and rea- combination of these, but patients with a grade 2 or 3 eruption sonable good activity against S. aureus. Agero et al., from required oral antibiotics (tetracyclines or cotrimoxazole) and the Dermatology Service of the Memorial Sloan, Kettering antihistamines (hydroxyzine). Cancer Center, suggest a treatment regimen with topical clin- damycin phosphate 1% gel for inflammatory pustules, or, if the gel is too irritating or a large body area is involved, 4. Suggestions by the Experts Panel oral tetracycline 250 mg four times daily or minocycline 100 mg two times daily [10]. They also suggest antihis- 4.1. Treatment tamines such as hydroxyzine to provide relief for patients with pruritus. Moreover, Agero et al. suggest to avoid the use Patients are advised to use mild skin cleansers, and use of topical antiacne medications such as retinoids and benzoyl sunscreens or decrease sun exposure if rash is aggravated peroxide since the reaction has a different pathophysiology by sunlight. In addition to sunscreens, alcohol-free emol- from acne vulgaris, aside from being drying and irritating lients are recommended (twice daily), also to reduce xerosis to the skin. As mentioned above, Lacouture suggests that and palliate itch. Emollients with a significant presence of clinical manifestations of EGFR inhibitors-related skin rash lipophilic components are to be preferred. The NCI-CTC can be staged chronologically (sensory disturbance-redness- categorization of skin rash (Table 1) is most commonly used oedema, papulopustular stage and crusting stage) [14].In to grade adverse events in clinical trials with HER1/EGFR- clinical practice, oncologists and dermatologists often treat targeted agents. In our opinion rash treatment, although the skin rash only in the papulopustular stage. On the contrary, empirical, should start from grade 2. In case of grade 1, the 160 C. Gridelli et al. / Critical Reviews in Oncology/Hematology 66 (2008) 155–162

Table 3 Table 4 Skin rash from erlotinib: treatment suggestions by the Experts Panel Skin rash from erlotinib: suggestions by the Experts Panel on dose reductions Grade 1 Topical treatment optional (sulfosalicylic creams, and treatment interruption topical antibiotics) Grade 1 No dose reduction or treatment interruption Grade 2 Topical treatment; if necessary add systemic Grade 2 No dose reduction or treatment interruption. In case of very tetracyclines and steroids; systemic antihistamines to distressing symptoms affecting quality of life if no response palliate pruritus to rash treatment, consider to interrupt erlotinib treatment Grade 3 Add to topical treatment systemic tetracyclines plus for 3–5 days and then re-start erlotinib at full-dose. If the systemic corticosteroids; antihistamines to palliate adverse event re-occur within 15 days consider to reduce pruritus; at resolution, for grade 3, consider prevention dose to 100 mg with short courses of tetracyclines Grade 3 Interrupt treatment until to improvement at least to grade 2. Grade 4 Refer to a burn-wound unit for intensive care Then re-start erlotinib at 100 mg/day. At re-occurrence also with reduced dose, interrupt definitively erlotinib treatment Grade 4 Interrupt definitively erlotinib therapy without trying opportunity to start a rash treatment is optional, and therefore dose-reduction should be discussed with the patient, in consideration of the low efficacy of empirical anti-rash therapies available. For grades 1 and 2 rash the panel suggested to use topical with full-dose without interruption. For grade 2 rash with treatment (i.e. 2% sulfosalicylic creams, topical antibiotics). very distressing symptoms affecting quality of life, and not In fact, the majority of the panelists referred empirical responding to treatment, short, but not excessively frequent, improvements with this sort of products, at least in the dose interruptions (3–5 days) can be considered and, if not mild-moderate reactions. Moreover, the panelists suggest to successful in improving quality of life of patients, with the consider the use of topical corticosteroids for grades 1–2 adverse event re-occurring within 15 days, dose reduction to rash, although clinical trials on their use are not available 100 mg/day is an option. For grade 3 rash we suggest to inter- and the attendees found their experiences controversial. In rupt treatment until to improvement at least to grade 2. Then fact, the panelists admitted that these topical agents (clobe- re-starting erlotinib at the reduced dose of 100 mg/day. At tasol, bethamethasone or triamcinolone) can be used without re-occurrence also with reduced dose we suggest to interrupt concern because skin atrophy occurs only after prolonged definitively erlotinib treatment. With grade 4 rash we sug- use and is reversible [35]. For more severe reactions (grade gest to interrupt definitively erlotinib therapy without trying 2 with more distressing symptoms and grade 3) the panel dose-reduction (Table 4). However, the panel suggests when suggested to add to topical treatment systemic antibiotics deciding about dose reduction/interruption to consider also (tetracyclines) and systemic corticosteroids. Intensive care the erlotinib activity and patient’s opinion. in a burn-wound unit is advised for grade 4 skin rash. Anti- histamines are recommended to provide relief for patients with pruritus. Recently, also pregabalin has been found to be 5. Suggestions for clinical research effective in the management of cetuximab-related itch [36]. The attendees suggested that topical retinoids should be fur- Obviously, a key recommendation from the Panel was to ther evaluated. In fact, although their skin-drying effects may begin controlled clinical trials of agents to treat rash. An exacerbate rash, anecdotal evidences of efficacy have been important suggestion for clinical research from the Panel was reported for these agents [35,37]. In fact, DeWitt et al. have to design clinical trials specific for the severity of skin rash. recently reported a case of dramatic clearance of skin rash The panel considered as research priorities the following with low-dose isotretinoin (20 mg daily) [35]. The panelists prospective comparisons of therapeutic approaches: topical also suggested to consider, in patients with severe and fre- treatment versus systemic corticosteroids, topical treatment quent episodes of rash, a prevention with short courses of versus systemic tetracyclines, and systemic tetracyclines ver- tetracyclines. In fact tetracyclines may be useful for their sus systemic corticosteroids. Moreover it is of interest to test weak anti-inflammatory effects and good activity against S. prospectively the efficacy of the prevention of skin rash with aureus, often responsible for secondary infection of the pus- short courses of tetracyclines. Also the activity of topical tules of these reactions (Table 3). At the moment the attendees immunomodulatory agents as pimecrolimus (Elidel®)isto discourage the use in clinical practice of topical immunomod- be tested in controlled clinical trials. ulatory agents as pimecrolimus (Elidel®) until clinical trials will not have tested their activity in skin rash from EGFR inhibitors. 6. Conclusions

4.2. Dose reduction and treatment interruption Skin rash is the most common adverse event associated with erlotinib treatment and it is often cause of negative Since rash treatment is quite unsatisfactory, the panel impact on the patients’ quality of life, particularly when gave some recommendations on drug interruption and dose- he is not well informed of. Moreover, this adverse event reduction. For grade 1 and grade 2 rash we suggest to continue may be cause of dose reductions, treatment interruption or C. Gridelli et al. / Critical Reviews in Oncology/Hematology 66 (2008) 155–162 161 discontinuation with a possible negative influence on the ther- [2] Parkin DM, Bray F, Ferlay J, Pisani P. Estimating the world cancer apeutic benefit of erlotinib. This is one of the reasons why burden: Globocan 2000. Int J Cancer 2001;94:153–6. the patients should be adviced to contact immediately the [3] Ciardiello F, Tortora G. Epidermal growth factor receptor (EGFR) as a target in cancer therapy: understanding the role of receptor expression oncologist in case of cutaneoeus side-effects. At the moment, and other molecular determinants that could influence the response to because of a lack of well-defined rash etiology and a lack of anti-EGFR drugs. Eur J Cancer 2003;39:1348–54. controlled clinical trials evaluating approaches to rash man- [4] Baselga J, Arteaga CL. Critical update and emerging trends in epi- agement, no evidence-based guidelines for skin rash treat- dermal growth factor receptor targeting in cancer. J Clin Oncol ment are available. In our opinion the recognition of the grade 2005;23:2445–59. [5] Cunningham D, Humblet Y,Siena S, et al. Cetuximab monotherapy and of the skin rash is very important for the management of this cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal side-effect. The NCI-CTC are to be used for the categoriza- cancer. N Engl J Med 2004;351:337–45. tion of the skin rash. Our Experts Panel suggests to consider [6] Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab an optional topical treatment for grade 1. A topical treatment for squamous-cell carcinoma of the head and neck. N Engl J Med is advised for grade 2, with the addition, if necessary, of a 2006;354:567–78. [7] Shepherd FA, Pereira JR, Ciuleanu T, et al. Erlotinib in previously systemic treatment with tetracylines and steroids. Systemic treated non small cell lung cancer. New Engl J Med 2005;353:113– tetracyclines and steroids are advised for grade 3; intensive 23. care in a burn-wound unit is advised for grade 4 skin rash. [8] Bezjak A, Tu D, Seymour L, et al. Symptom improvement in lung cancer Our Experts Panel gives great relevance to dose reduction patients treated with erlotinib: quality of life analysis of the National and interruption. For grades 1–2 rash we suggest to continue Cancer Institute of Canada Clinical Trials Group Study BR.21. J Clin Oncol 2006;24:3831–7. with full-dose without interruption. However, for grade 2 with [9] Robert C, Soria JC, Spatz A, et al. Cutaneous side-effects of very distressing symptoms affecting quality of life, short, but kinase inhibitors and blocking antibodies. Lancet Oncol 2005;6:491– not excessively frequent, dose interruptions (3–5 days) can 500. be considered and, if not successful in improving quality of [10] Agero AL, Dusza SW, Benvenuto-Andrade C, Busam KJ, Myskowski life of patients, dose reduction to 100 mg/day is an option. P, Halpern AC. Dermatologic side effects associated with th epi- dermal growth factor receptor inhibitors. J Am Acad Dermatol For grade 3 rash we suggest directly to reduce erlotinib dose 2006;55:657–70. at 100 mg/day. With grade 4 rash we suggest to interrupt [11] Soler RP, Delord JP, Halpern A, Kelly K, Krueger J, Sureda BM. definitively erlotinib therapy without trying dose-reduction. HER1/EGFR inhibitor-associated rash: future directions for manage- The increasing use of erlotinib in NSCLC treatment, the ment and investigation outcomes from the HER1/EGFR inhibitor rash development of new EGFR inhibitors and simultaneously management forum. Oncologist 2005;10:345– 56. [12] Busam KJ, Capodieci P, Motzer R, et al. Cutaneous side-effects in the lack of clinical trials on this topic, makes prospective cancer patients treated with the antiepidermal growth factor receptor investigation on rash management mandatory. antibody C225. Br J Dermatol 2001;144:1169– 76. [13] Jacot W, Bassis D, Jorda E, et al. induced by epi- dermal growth factor receptor inhibitors in patients with solid tumors. Reviewers Br J Dermatol 2004;151:238–41. [14] Lacouture ME. Mechanisms of cutaneous toxicities to EGFR inhibitors. Rafael Rosell, MD, PhD, Medical Oncology Service, Nat Rev Cancer 2006;6:803–12. Catalan Institute of Oncology, Hospital Germans Trias i Pujol [15] Johnson JR, Cohen M, Sridhara R, et al. Approval summary for erlotinib Ctra Canyet, s/n Badalona, Barcelona E-08916, Spain. for treatment of patients with locally advanced or metastatic non-small cell lung cancer after failure of at least one prior chemotherapy regimen. Mario E. Lacouture, MD, Assistant Professor, Department Clin Cancer Res 2005;11:6414–21. of Dermatology Robert H Lurie Comprehensive Cancer Cen- [16] Roskos L, Lohner M, Osborn K, et al. Low pharmacokinetic variability ter, Northwestern University, 676 North St. Clair Suite, 1600 facilitates optimal dosing of ABX-EGF in cancer patients. Proc Am Chicago, IL 60611-2941, United States. Soc Clin Oncol 2002;21:1 [abstract 362]. Jacques Bernier, MD, PhD, Dept. of Radio Oncology, [17] Laux I, Jain A, Singh S, Agus DB. Epidermal growth factor receptor dimerization status determines skin toxicity to HER-kinase targeted Clinique de Genolier, 1 route du Muids, Genolier 1272, therapies. Br J Cancer 2006;94:85–92. Switzerland. [18] Perea S. Genotypic bases of EGFR inhibitors pharmacological actions. J Clin Oncol 2004;22:3005. [19] Hidalgo M, Siu LL, Nemunaitis J, et al. Phase I and pharmacological Conflicts of interest study of OSI-774, an epidermal growth factor receptor tyrosine kinase inhibitor, in patients with advanced solid malignancies. J Clin Oncol 2001;19:3267–79. Roche sponsored this Experts Panel Meeting, and all the [20] Vallbohmer D, Zhang W, Gordon M, et al. Molecular determinants of authors of the present manuscript received honoraria for par- cetuximab efficacy. J Clin Oncol 2005;23:3536–44. ticipating to the Meeting. [21] Perez-Soler R. Can rash associated with HER1/EGFR inhibition be used as a marker of treatment outcome? Oncology (Huntingt) 2003;17:23–8. [22] Perez-Soler R, Chachoua A, Hammond LA, et al. Determinants of References tumor response and survival with erlotinib in patients with non-small- cell lung cancer. J Clin Oncol 2004;22:3238–47. [1] Jemal A, Siegel R, Ward E, Murray T, Xu J, Thun MJ. Cancer statistics [23] Perez-Soler R, Saltz L. Cutaneous adverse effects with HER1/EGFR- 2007. CA Cancer J Clin 2007;57:43–66. targeted agents: is there a silver lining? J Clin Oncol 2005;23:5235–46. 162 C. Gridelli et al. / Critical Reviews in Oncology/Hematology 66 (2008) 155–162

[24] Saltz LB, Kies MS, Abbruzzese JL, Azarnia N, Needle M. The pres- non-small cell lung cancer (NSCLC): surprising efficacy of early local ence and intensity of the cetuximab-induced acne-like rash predicts treatment. J Clin Oncol 2004;22(Suppl):14S [abstract 7100]. increased survival in studies across multiple malignancies. Proc Am [35] DeWitt CA, Siroy AE, Stone SP. Acneifomr eruptions associated with Soc Clin Oncol 2003;22:204 [abstract 817]. epidermal growth factor receptor-targeted chemotherapy. J Am Acad [25] OSI Pharmaceuticals. Tarceva package insert; 2005. Dermatol 2007;56:500–5. [26] ImClone Systems. Erbitux package insert; 2006. [36] Porzio G, Aielli F, Verna L, et al. Efficacy of pregabalin in the [27] Roe` E, Garcia Muret MP, Marcuello E, et al. Description and man- management of cetuximab-related itch. J Pain Symptom Manage agement of cutaneous side effects during cetuximab or erlotinib 2006;32:397–8. treatments: a prospective study of 30 patients. J Am Acad Dermatol [37] Cowen EW. Epidermal growth factor receptor inhibitors: a new era of 2006;55:429–37. drug reactions in a new era of cancer therapy. J Am Acad Dermatol [28] Galimont-Collen AFS, Vos LE, Lavrijsen APM, Ouwerkerk J, 2007;56:514–7. Gelderblom H. Classification and management of skin, hair, nail and mucosal side-effects of epidermal growth factor receptor (EGFR inhibitors). Eur J Cancer 2007;43:845–51. [29] Mak KK, Chan SY. Epidermal growth factor as a biologic switch in Biography hair growth cycle. J Biol Chem 2003;278:26120– 6. [30] Jacot W, Bessis D, Jorda E, et al. Acneiform eruption induced by epi- Cesare Gridelli, M.D., is currently Chief of Division of dermal growth factor receptor inhibitors in patients with solid tumors. Medical Oncology and Director of Department of Oncol- Br J Dermatol 2004;151:238–41. ogy/Hematology at the “S.G. Moscati” Hospital of Avellino [31] Klmyal-Asadl A, Jlh MH. Follicular toxic effects of chimeric anti- epidermal growth factor receptor anti-epidermal growth factor receptor (Italy). His areas of expertise are lung cancer, cancer in the antibody cetuximab used to treat human solid tumors. Arch Dermatol elderly, antiemetics. He is involved in the clinical develop- 2002;138:129–31. ment of new anticancer agents. He is member of Advisory [32] Herbst RS, LoRusso PM, Purdom M, WardD. Dermatologic side effects Board of scientific journals and of several expert panels. associated with gefitinib therapy. Clinical experience and management. Dr. Gridelli has been invited speaker of international con- Clin Lung Cancer 2003;4:366–9. [33] Shah NT, Kris MG, Pao W, et al. Practical management of patients ferences and educational activities of oncology societies with non-small-cell lung cancer treated with gefitinib. J Clin Oncol (ASCO, ESMO, etc.). He is author or co-author of about 2005;23:165–74. 450 papers, of which about 180 extended papers published on [34] Cortesi E, De Pasquale Ceratti A, D’Auria G, et al. Management of international indexed journals, and several chapters of books. cutaneous adverse effects during treatment with ZD1839 in advanced