Contemporary, CRC is considered a molecularly heterogeneous malignancy characterized by several genomic and epigenomic alterations used to assess prognostic information and to guide the best treatment approach [21, 22]. Targeted therapies, including anti-EGFR panitumumab and cetuximab, have been introduced in the treatment of mCRC at selective settings [22–28]. Although OAEs related to panitumumab have been reported in mCRC patients, it lacked well-documented descriptions of these cases as well as an analysis of effectiveness on the management of these events [2, 5, 6]. To the best of our knowledge, we present the first study documenting the clinical patterns of aphthous-like and atrophic tongue lesions associated with panitumumab-containing regimens and the outcomes following our experience in treating this condition.
A systematic review that analyzed the risk of grade ≥ 3 mucositis in CRC patients treated with anti-EGFR regimens found that the risk ratio was 3.44, particularly for patients undergoing panitumumab-containing regimes . Besides the apparent low risk of severe cases (grade ≥ 3) of mucositis/stomatitis for patients undergoing panitumumab in accordance with our findings, the impact on patients’ quality of life might be underestimated, considering the frequency of modifications needs on ongoing treatments and patient-reported symptoms described in this series. Additionally, there has been no standard characterization of this phenomenon in the available trials, and experienced oral clinicians are not usually involved in these studies. In a brief review of studies regarding panitumumab safety analysis, a broad range of terms such as mucositis, stomatitis, mucosal inflammation, among other unspecific terms, were included to reporting oral mucosa changes [22, 25–29], which may compromise the accuracy in systematically determine whether they represented conventional oral mucositis or were in the medication-associated stomatitis spectrum. The marked aphthous-like appearance and tongue atrophic lesions observed in our study suggest that these cases represent more likely stomatitis than genuine oral mucositis as the cases described by Sonis et al. previously . Therefore, we believe that the term panitumumab-associated stomatitis may better represent the clinical profile of this oral toxicity.
Most commonly, anti-EGFR targeted therapies have been associated with a significantly increased risk of skin toxicity, affecting even 90% of the patients undergoing these drugs [30–33]. Dermatologic changes associated with panitumumab include a broad spectrum that may involve cutaneous and skin adnexal disorders [32, 33]. The patients analyzed in the current study presented several dermatologic toxicities during the treatment course. Acneiform rash in the face and/or trunk was present in all our patients attending concomitantly to oral lesions. It frequently represented an early event of clinical relevance in the toxicity profile and preceded the oral lesions. The close biological and molecular signature of mucosal and epidermal keratinocytes may explain this specific profile which reflects mainly disturbances of EGFR inhibition. Interestingly, severity and early timing onset of skin toxicity has been associated with better survival outcomes, being a surrogate marker of a better panitumumab clinical response [31, 33]. Considering the limited number of cases presented herein, it is not already possible to assume the same conclusions regarding the oral toxicity profile.
EGFR activation is involved in several physiologic cellular processes as differentiation, proliferation, cell migration, and blood flow changes, essential for mucocutaneous homeostasis and tissue repair. EGF signaling is considered a key regulator of gastrointestinal tract mucosal barrier integrity , and disturbances in this pathway are associated with increased epithelium permeability and wound healing impairment [35, 36]. EGF that is constitutively produced in salivary fluid has been investigated in the pathophysiology of recurrent aphthous stomatitis, even though some results were controversial [37, 38]. Gu et al.  suggested that the decrease in salivary EGF in the remission interval of patients with recurrent aphthous stomatitis would be related to the ulcer initiation phase, and EGFR activation may be crucial to the resolution of the aphthous lesions. However, another study by Rezae et al.  did not found differences in salivary EGF levels between affected patients and controls. Recurrent aphthous stomatitis, which closely resembles the oral lesions observed herein, is considered an immunologically driven multifactorial disease of unknown etiology [41, 42] and the current findings regarding EGF-EGFR disturbances may add a valuable information to better understand its intriguing and complex pathogenesis.
Although the underlying pathogenesis of panitumumab-associated stomatitis remains unclear at this point, we hypothesize that the inhibition of EGFR may elicit a particular inflammatory and apoptotic signaling to the oral mucosa cells, predisposing to epithelial damage and disruption of the mucosal barrier, which ultimately can result in ulceration and pain in a similar way to recurrent RAS lesions. Additionally, the complex EGFR inhibitory effects on the immune system response and the decrease in the immunosurveillance to oral pathogens can also be involved in this process. Pathogens such as HSV and the new coronavirus (SARS-CoV-2) were associated with increased severity in our cases.
Another point is that the combination of panitumumab to chemotherapy and the type of concurrent drug may contribute to the compromise of the protective factors and regenerative capacity of the oral mucosa. Notwithstanding, most of our cases were previously treated with irinotecan alone before starting the combined therapy with panitumumab in a subsequent line, and no evidence of reported oral lesions was retrieved in our analysis, which point out to the key role of the anti-EGFR effects. In addition, the clinical presentation of our cases was similar regardless of concurrent chemotherapeutic scheme or treatment line, but the treatment combinations did not vary significantly across the patients. Interestingly, anti-EGFR tyrosine kinases inhibitors used to treat advanced non-small-cell lung cancer have been attended to a similar risk of stomatitis [43, 44], which corroborate our idea that EGFR blockage effects seem pivotal on pathogenic events following this OAE.
Although delaying on treatment sequence, interruptions or even discontinuation due to oral lesions represented few cases, our results emphasize the emerging need of guided oral care protocols for mCRC patients undergoing anti-EGFR targeted drugs. Despite the small sample size, the herein characterized panitumumab-associated stomatitis responded well either to topical corticosteroids or to PBM alone or in combination, which improved the healing of impaired EGFR-signaling oral mucosa cells. We also believe that basic oral care and oral palliative treatment may be incorporated to the routine of treatment of patients that develop these mouth sores. Even though PBM proved to be effective as well, the compliance to daily applications and the variability of possible settings may be taken in count. Preventive measures including light-based approaches and systemic or topical corticosteroids seem to be encouraging.
Based on the retrospective nature of our study, we highlight some limitations of our findings. The small number of patients by the convenience of the sample, the lack of heterogeneity between the treatment protocols which did not include patients undergoing panitumumab as monotherapy, and the variability in the management of this condition that was not uniform across the patients and during the patients’ follow-up are some examples that would be overcome in the future. Prospective studies, including clinical trials, will be necessary to design preventive strategies and test curative protocols in managing panitumumab-associated stomatitis more accurately.
In conclusion, panitumumab-containing regimes were associated with a particular pattern of painful aphthous-like and atrophic tongue lesions that were accompanied by a marked acneiform cutaneous rash following the beginning of the therapy as well as several other AEs. The lesions were similar to recurrent aphthous stomatitis, and it was eventually able to cause modifications or discontinuation of the antineoplastic treatment schedule like the previously reported mIAS. Our preliminary results found that topical corticosteroids as clobetasol propionate 0.05% and/or PBM were effective on treating panitumumab-associated stomatitis. The anti-EGFR effects of panitumumab on oral mucosa cells homeostasis seem to play a central role in the pathogenesis of this poorly understood condition. We strongly believe that preventive measures, early recognition by trained clinicians, and the correct management of this OAE may be helpful to improve the compliance of patients to the antineoplastic treatment and maintain their quality of life at adequate levels during the treatment course of panitumumab.