Management of severe bio-radiation dermatitis induced by radiotherapy and cetuximab in patients with head and neck cancer: emphasizing the role of calcium alginate dressings

Cetuximab (CTX) is a chimeric mouse IgG1 monoclonal anti-epidermal growth factor receptor antibody approved for use in patients with metastatic colorectal and head and neck cancers. Since 2006, the combination of radiotherapy (RT) and CTX is an evidence-based treatment option [1] for the curative treatment of locally advanced head and neck squamous cell carcinoma (HNSCC). It has long been known that the acute skin effects induced by radiation are dose-, volume-, and treatment-time dependent [2]. No topical interventions have been demonstrated to prevent [3] the occurrence of radiation dermatitis in head and neck cancer patients. Nonetheless, severe reactions are relatively uncommon when Intensity-Modulated Radiotherapy (IMRT) is employed, thanks to its ability to avoid [4] excessive superficial dose accumulation. The concomitant use of CTX and RT in HNSCC is frequently associated with the development of overlapping “in-field” side effects, generally defined as “bio-radiation” dermatitis (BRD) [5, 6]. Compared with what can be observed with RT alone, BRD is characterized by peculiar clinical features which can present already in the first 2 weeks of treatment: an enhanced inflammatory response with coexistence of desquamation and acneifom rash can be associated with the development of confluent crusts, diffuse xerosis, and intense pruritus within the affected area. The severity of BRD can result in significant pain, impairment of quality of life [7], and predisposition to infections [8] and sepsis [9]. After the approval of CTX-RT in HNSCC, several authors [10‐12] questioned the feasibility of the concomitant treatment due to excessive skin toxicity and risk of treatment breaks, a known detrimental factor [13] to the efficacy of RT. In order to address these issues, a series of consensus guidelines were published [5, 6, 14‐16], providing expert recommendations on the optimal recognition and management of cutaneous side effects induced by CTX-RT. However, the impact of grade-specific algorithms on daily practice, as presented in these articles, has never been investigated in prospective trials [17]. In addition, the proposed strategies are not consistent regarding what should be the best approach required to treat severe reactions. In 2014, an advanced wound care nursing team was established in our clinic to prospectively monitor and manage the development of BRD in patients with HNSCC undergoing concurrent RT-CTX. From a therapeutic standpoint, based on our experience, we advocate the central role of calcium alginate fibers. Being highly absorbent dressings [18], these products are able to remove excessive exudate while maintaining adequate moisture in the affected tissue. Thus, they represent a very effective treatment option to deal with the extensive, multifactorial moist desquamation frequently observed in these cases.

The aim of our work was to assess the impact of an in-house skin policy protocol on the tolerability of concurrent RT and CTX in patients with HNSCC. In addition, we investigated whether the routine availability of an advanced wound care nursing team played a role in preventing the development of severe BRD.

Between February 2007 and January 2018, 51 patients were treated with RT-CTX for HNSCC. Patients’ characteristics are summarized in Table 1. The median age of the cohort was 65 years (range 44–81). Most patients were burdened with multiple comorbidities, with an overall median CCI of 6 at cancer diagnosis. All were deemed cisplatin-unsuitable whereas almost a fourth had very advanced disease (stage IVB disease in 23.5% of cases). A sequential approach based on induction chemotherapy followed by RT-CTX was prescribed only for selected patients (15.7%). All patients received curatively intended radiation, in most cases (66.7%) delivered with modern IMRT technique. Treatment features and compliance data are shown in Table 2. Regarding acute toxicity, the rates of G3/G4 mucositis, BRD, dysphagia, and acneiform rash were 54.9%, 43.1%, 31.3%, and 13.7%, respectively. Sepsis developed in 4/51 patients (7.8%), in one of them during the fourth week of RT-CTX. Altogether, a permanent discontinuation of treatment before reaching a curative dose of 66 Gy was observed in three cases because of sepsis, massive tumor bleeding, and patient’s refusal, respectively. In terms of patients’ tolerability, a prolonged RT treatment break (> 7 days) was required for about one third of the whole cohort (35.3%), mainly due to excessive acute toxicity or worsened general conditions. In order to assess whether the routine availability of an advanced wound care nursing team played a role in preventing the development of severe skin toxicity, two groups of patients were identified: 26 subjects treated between February 2007 and November 2013 were compared with 25 patients who underwent RT-CT between March 2014 and January 2018. In the latter group, two patients had G3 infusional reaction at first CTX introduction and were therefore not considered for toxicity analysis. The incidence of G3/G4 mucositis, BRD, dysphagia, and acneiform rash were 73% (19/26), 38.4% (10/26), 53.8% (14/26), 7.6% (2/26) in the first group and 39.3% (9/23), 47.8% (11/23), 8.6% (2/23), and 21.7% (5/23) in the second group.

No significant differences were therefore found between the two groups in terms of G3/G4 BRD and acneiform rash (p = 0.65 and p = 0.9, respectively) nor in terms of mucositis and dysphagia (p = 0.48 and 0.46, respectively). Overall, only posttreatment relative weight loss > 10% compared with baseline correlated significantly with the development of G3/G4 BRD in the whole cohort (39% vs 18% if weight loss below 10%; p = 0.013).

In terms of treatment tolerability, the implementation of a standardized in-house skin protocol by January 2014 was correlated with a significant improvement of both RT and CTX compliance. In particular, a mean RT interruption of 8.42 days (SD 6.73; 95% CI 5.7–11.1; median 10 days) was reported in the older group vs 0.86 days in the more recent cohort (SD 2.66; 95% CI − 0.28–2.02, median 0 days; p < 0.0001) (Fig. 1). Notably, the adoption of our standardized skin management protocol allowed to complete the treatment without any interruption in 20/23 patients (89.6%) vs only 6/26 (23%) in the older cohort. At univariate analysis, also the use of IMRT (p = 0.0033) was correlated with better tolerability; however, at multivariate analysis, only the absence of a standardized supportive approach was independently associated with RT interruptions longer than 7 days (OR 23.4, 95% CI 4.3–125.6, p = 0.0003). In addition, a higher mean CTX RDI (86.3% vs 74.5%) was significantly correlated only with the use of our skin toxicity protocol at univariate analysis (p = 0.0226) (Fig. 2). At time of analysis, 47% of patients are alive. At a median follow-up of 13 months (range, 0–133), the 1- and 2- year rates of LRC, PFS, and OS were 59% and 55.5%, 50% and 37.5%, and 65% and 44%, respectively (Fig. 3 and supplementary material).

The exact incidence of the development of G3/G4 BRD in head and neck cancer is a matter of debate. A large variability can be observed by analyzing phase 3 and randomized phase 2 trials where a combination of RT and CTX was adopted, either alone or as backbone of concurrent treatment (Table 3). Despite the fact that expert consensus recommendations were published with the aim to guide supportive skin care, the impact of grade-specific management on the tolerability of RT-CTX was not assessed in these large prospective studies. Notably, a multicenter randomized double-blind trial [22] failed to demonstrate the efficacy of a topical glycosaminoglycan analog (OTD70-DERM) over placebo in reducing the incidence and severity of BRD. On a broader scale, no high-level evidence [23] supports the adoption of any specific measures to prevent radiation dermatitis, nor of therapeutic interventions to manage it when RT alone is employed, as well. Identifying the best approach to deal with severe cutaneous reactions poses therefore a significant challenge in head and neck cancer curative treatment, pending the need to avoid treatment breaks. A large multi-cohort randomized study on 357 patients [24] was designed to compare the efficacy of a hydrogel (Intrasite) and a dry dressing (Tricotex) for moist desquamation. Most common primary tumor site was breast (63.3%) followed by HNSCC (28.8%). The results of this study did not warrant the use of gel dressings for severe reactions induced by RT, since their use was associated with a significantly prolonged time to healing (HR, 0.64; 95% CI, 0.42–0.99). In breast cancer patients, the prophylactic use of a dry dressing (Mepitel film) was shown [25] to be very effective in preventing moist desquamation, reducing overall skin reaction severity by 92% compared with standard aqueous cream. The same group performed a feasibility study [26] on the application of Mepitel Film in a cohort of 36 head and neck cancer patients undergoing standard cisplatin-based chemoradiation. In comparison with standard moisturizers, the incidence of moist desquamation was reduced by 37% and 28% in subjects treated in China and New Zealand, respectively. Of note, the ongoing Rarest-01 randomized trial [27] was planned to demonstrate a reduction of > G2 radiation dermatitis from 85 to 65% through the preventive use of Mepitel Film versus standard management in a cohort of 168 HNSCC patients. With the exception of the GORTEC 2009-01 trial [22], the adoption of supportive measures for skin toxicity was not tested in HNSCC patients treated with concurrent RT and CTX. Taking into account the underlying biologic interplay between the EGFR pathway and molecular effects of radiation [28], ad-hoc supportive treatments have to be designed to counterbalance the marked inflammatory response commonly observed in the irradiated field when CTX is administered on top of radiation. In our experience, the use of a standardized, in-house protocol for the management of BRD was an effective strategy to increase patients’ tolerability to RT-CTX and, importantly, to crucially minimize treatment breaks. Heterogeneity in skin care among different RT departments [29] may be particularly relevant when dealing with overlapping/complex side effects such as those induced by RT and CTX. The early employment of calcium alginate fibers to handle moist desquamation was a key feature of our wound care approach. Unlike previous recommendations [5, 6], the use of topical dressings was anticipated from G3 to G2 BRD. Another relevant difference from consensus publications [15, 16] regards the role of hydrocolloid agents. In contrast to them, their role was much less relevant in our routine practice. Essentially, their use was limited to wound care taking place after the end of RT or as secondary dressings to conform the primary fiber to head and neck anatomy. Several limitations have to be acknowledged in interpreting our data. First, the retrospective nature of the study and its small sample size restrain the generalizability of our findings. Second, the absence of a control group, being it patients treated with RT alone or cisplatin-based chemo-radiation, does not allow to draw firm conclusions on how large is the benefit of our skin policy on treatment tolerability. Third, the need for daily dressing changes as described in our protocol can be labor intensive and discomforting for the patients requiring them. When dealing with skin toxicity induced by anti-EGFR treatment, consideration should be given to patients’ preferences [30, 31] also regarding the potential need of lengthy supportive care. In this perspective, quality of life measures were not formally evaluated in our study. In line with previously published studies [17], > G3 BRD was not negligible in our cohort. As recommended by expert consensus recommendations, our data underline the importance of tailored supportive care interventions for the management of “in-field” skin toxicity induced by concurrent RT and CTX, allowing for a marked improvement of compliance to treatment. In light of the paucity of data available on this topic, our experience may be regarded as hypothesis-generating in relation to the central role of alginate dressings we described.

The combination of RT and CTX is a possible treatment option for curative treatment of HNSCC in patients deemed unfit for cisplatin-based chemoradiation. In order to maximize treatment efficacy, patients’ tolerability to both therapies should be assured by managing the frequent occurrence of severe bio-radiation dermatitis through standardized, grade-specific interventions. For this purpose, routine involvement of an advanced wound care management team and early application of calcium alginate dressings for moist desquamation are warranted.