The Effects of Proton and Photon Radiation Therapy on the Development of Pediatric Dermatitis

Simple Summary

Radiation therapy is a common and effective cancer treatment, but it frequently causes a skin reaction known as radiation-induced dermatitis. This study aims to assess the frequency and severity of radiation-induced dermatitis in children receiving two types of radiation therapy—proton and photon treatments. By reviewing the medical records of pediatric cancer patients treated between 2018 and 2023 at a single institution, the study examined factors such as total radiation dose and the addition of other oncological treatments to determine their impact on the severity of dermatitis. The findings revealed that higher radiation doses and concurrent chemotherapy were associated with more severe cases of dermatitis, while no significant difference was observed between the two radiation types. These results may assist healthcare providers in optimizing treatment plans for pediatric cancer patients by considering adjustments to radiation dosage or what therapies to combine to minimize side effects like dermatitis.

Abstract

Although radiation therapy is the leading option for effective cancer treatment, a prevalent side effect associated with it is dermatitis. Despite some available literature on this topic, there remain many gaps that need to be addressed. The goal of this study is to determine the incidence of radiation-induced dermatitis (RID) among children receiving proton and photon therapies; a retrospective chart review, at a single institution, was conducted on oncology patients who underwent proton or photon therapy radiation between 2018 and 2023. Significant differences were found between the Radiation Therapy Oncology Group (RTOG) score and the total radiation dose (p = 0.04). The median total dose of radiation received by those with an RTOG score of l was 5040.0 mGy and increased to 7600 mGy for those with a score of 3. A significant association was found between those who received chemotherapy and dermatitis (p = 0.04). No significance was found between the incidence of dermatitis in photon and proton therapy (p = 1.00). The study showed that multiple factors, including total radiation dose and chemotherapy, can affect RID. These relationships can be used to determine the modality, dose, and additional treatment options best suited to treat cancer patients in the pediatric population.

1. Introduction

Despite the success of radiation therapy treating patients with malignancies, radiation-induced dermatitis (RID) remains a common side effect that negatively impacts patient quality of life. To improve patient care, it is important to explore the prevalence, risk factors, and treatment options that are associated with dermatitis in radiation therapy.

Photon therapy involves using high-energy X-rays to deliver radiation [1]. Since photons are massless, the radiation is delivered through continuous energy deposit and gradual dissipation of the energy to the tissues [2]. Within the commonly reported side effects of photon therapy, dermatitis is noted as an issue that could affect treatment continuity [3]. On the other hand, proton therapy, using positively charged particles, allows for the precise targeting of tumor masses through the Bragg peak phenomenon, minimizing damage to the surrounding healthy tissue [1]. The protons deposit energy preferentially inside the tumor, reducing radiation-induced side effects compared to traditional methods [2]. Among children, proton therapy is used to target tumors located near important structures and radio-resistant tumors [2]. Although photon therapy remains the traditional and primary radiation modality with impressive clinical outcomes, the localized dosing of protons makes it a more appealing option [1]. Another perceived benefit in proton therapy is from diminishing the volume of exposure to radiation, thereby potentially minimizing long-term adverse effects. This can be particularly important among children who will likely be long-term survivors [4]. Early studies show that proton therapy could help minimize the severity of acute skin toxicity in children receiving radiation therapy [5].

Dermatitis in radiation therapy can develop due to intrinsic factors such as genetic predisposition, pre-existing skin conditions and extrinsic factors like radiation dose, mode of treatment, treatment time, the amount of radiated surfaces and concurrent therapies [6]. The likelihood of developing radiation-induced dermatitis increases with higher doses of radiation over prolonged periods [7]. Symptoms of dermatitis include erythema, dryness, itching, desquamation, edema, and pain, and are managed using topical emollients, corticosteroid creams, dressing changes, and cooling agents [8,9]. The severity of the dermatitis can be measured through grading systems such as the RTOG Acute Radiation Morbidity Scoring Criteria. RTOG scores are assigned with grading classifications ranging from zero to five, with zero considered normal and five considered so severe it leads to death.

In pediatric patients, additional factors like pediatric skin sensitivity, differences in the location of anatomical structures, tissue composition, and interactions with growth and development can influence the development of RID [10]. Furthermore, the psychological and emotional state of patients can affect susceptibility of developing dermatitis. These features need to be taken into account when devising tailored interventions and personalized care strategies for pediatric patients.

A systematic review by Gamvroula et al. (2022) [6] found wide variation in the reported prevalence of radio-dermatitis among pediatric, adolescent, and young adult cancer patients, with the prevalence of grade 3 dermatitis ranging from 6% to 40% in proton therapy patients and 5% to 80% in photon therapy patients. However, inconsistent methodologies, often relying on clinical observations and different grading systems, contribute to varying results. Some studies noted a lower incidence of dermatitis in photon therapy compared to proton therapy like Pasalic et al. (2020) [8] when observing patients with testicular seminoma. Meanwhile, others, like Gaito et al. (2021) [1] and Grant et al. (2015) [11], observed no difference between the two modalities. Further research is needed on long-term radiation-induced skin injuries [12].

It is important to acknowledge the limitations in these studies, including sample size, differences in treatment protocols, patient characteristics, and dose distributions which may add to the inconsistency in observed results. Additionally, the retrospective nature of these studies could potentially introduce biases, making it challenging to generalize findings. Addressing these limitations through future research that utilizes standardized methodologies and incorporates larger sample sizes allows for more definitive conclusions when discussing the prevalence of dermatitis in proton therapy and photon therapy.

In this study, we hope to address some of these limitations by focusing specifically on the differences in incidence of dermatitis between proton and photon therapy within the pediatric population. By limiting the study to only look at characteristics of skin toxicity we hope to definitively understand whether there is a difference in dermatitis between the two modalities. We plan to use a standardized approach in determining dermatitis that can be replicated in future studies.

2. Materials and Methods

At our institution, we performed a retrospective chart review and identified patients receiving cancer treatment at Cook Children’s Hospital between 2018 and 2023. Patients who had radiation data in the Cancer Registry and those who had information on radiation treatment in patient charts were further down-selected.

These patients were categorized on whether they received photon or proton therapy. A total of 134 patients received proton therapy, 13 patients underwent photon therapy, and 1 patient received a combination of both therapies. Radiation treatment summaries were reviewed to determine the total radiation dose, number of fractions and radiation type for each patient. A chart review was conducted to identify whether patients received any surgery or chemotherapy as part of their treatment plan and the specific location of the radiation therapy. Subsequent patient visits were examined after therapy for any mention of radiation-induced dermatitis. Those conditions were then tracked to assess the severity, characteristics of the dermatitis, the treatments administered, and the time to resolution. RTOG scores were assigned with grading classifications ranging from 0 to 5. Zero was considered normal, one had characteristics of slight dermal atrophy with faint erythema, two had tender and bright red erythema with patchy desquamation, three had confluent desquamation with pitting edema, four showed signs of ulceration, and five meant death.

Descriptive statistics and frequencies were used to calculate the median, interquartile range (IQR), and distribution of categorical variables through IBM SPSS v25.0 software. The primary objective was assessed using Fisher’s Exact Test and the likelihood ratio test to determine if there was an association between therapy type and the incidence of dermatitis. Due to limitations in sample size, statistical tests for the secondary objectives were utilized for exploratory purposes. A Kruskal–Wallis test was used to compare total dose and fractions between RTOG scores. The likelihood ratio statistic was then used to determine if there was an association between the incidence of dermatitis and patients who received chemotherapy, surgery, or both. Finally, a Mann–Whitney U test was used to determine if there were any differences in time to resolution between those who received intervention and those who did not. All statistical tests were computed with an alpha level of 0.05.

3. Results

3.1. Descriptive Statistics

A total of 172 patients were identified with radiation treatment data from the 5 year range and 24 patients were excluded from analysis due to missing data. The missing data included information on total dose, number of fractions, and radiation type. Among these 148 patients, age at treatment ranged from <1 to 32 years old, with a median of 10.5. Diagnoses included but were not limited to rhabdomyosarcoma, ependymoma, Hodgkin lymphoma, Ewing sarcoma, medulloblastoma, pilocytic astrocytoma, neuroblastoma, and Wilms tumor. The majority of the sample was male (64%), White (78%), and not Hispanic or Latino (71%). The median total dose and fraction was 5220.0 and 28.0. In addition, head and neck cancer was the most prevalent (60%), and most patients had both chemotherapy and surgery (71%), and no reported toxicities (82%).

Table 1. Provides the demographics of the sample, broken down into therapy type, and then the total for the whole sample. Age, total dose, and fraction are presented as a median and IQR. The categories are presented in bolded terms.

Table 1	Proton (n = 134)	Photon (n = 13)	Both (n = 1)	Total (n = 148)
Gender
Male	88 (66%)	7 (54%)	0 (0%)	95 (64%)
Female	46 (34%)	6 (46%)	1 (100%)	53 (36%)
Race
American Indian/Alaska Native	1 (1%)	0 (0%)	0 (0%)	1 (1%)
Asian	4 (3%)	0 (0%)	0 (0%)	4 (3%)
Black or African American	22 (16%)	2 (15%)	0 (0%)	24 (16%)
White	104 (77%)	10 (77%)	1 (100%)	115 (78%)
More Than One Race	1 (1%)	0 (0%)	0 (0%)	1 (1%)
Unknown/Not Reported	2 (2%)	1 (8%)	0 (0%)	3 (2%)
Ethnicity
Hispanic or Latino	36 (27%)	5 (38%)	1 (100%)	42 (28%)
NOT Hispanic or Latino	97 (72%)	8 (62%)	0 (0%)	105 (71%)
Unknown/Not Reported	1 (1%)	0 (0%)	0 (0%)	1 (1%)
Primary Site
Head and Neck	81 (60%)	8 (62%)	0 (0%)	89 (60%)
Trunk	36 (27%)	2 (15%)	0 (0%)	38 (26%)
Limbs	15 (11%)	2 (15%)	1 (100%)	18 (12%)
Unknown	2 (2%)	1 (8%)	0 (0%)	3 (2%)
Age (Median IQR)	11.0 (6.0–15.0)	6.0 (4.5–10.5)	3.0	10.5 (6.0–15.0)
Total Dose	5220.0	4350.0		5220.0
(Median IQR)	(3600.0–5715.0)	(1200.0–5805.0)	1080.0	(3600.0–5670.0)
Fraction (Median IQR)	29.0 (19.0–31.3)	12.0 (7.0–30.0)	6	28.0 (14.0–31.0)

provides more detail of descriptive statistics, divided by therapy type. The individual who received both therapies was a female of white and Hispanic descent who was 3 years of age, receiving a total dose of 1080 mGy and 6 fractions. Of all the patients who received radiation, 9% of the sample had atopic dermatitis, 7% specifically had diaper dermatitis, 1% had folliculitis, 1% had hidradenitis suppurativa, 2% had impetigo, 1% had paronychia, 1% had post-chemotherapy folliculitis and 1% had stoma dermatitis. A total of 23% of the sample had some form of dermatological history prior to receiving radiation.

Only 8% of patients receiving proton therapy underwent additional chemotherapy alone, 18% received surgery alone, and 69% received both treatments. In the photon therapy group, 85% received both chemotherapy and surgery. Acute toxicity was observed in 13% of the proton therapy group, with late toxicity occurring in 5%, whereas both acute and late toxicities were noted in 8% of the photon therapy group. Among those receiving proton therapy, 10% required an intervention, compared to 15% in the photon group. Comorbidities were seen in 19% of the proton group at 31% of the photon group. Lastly, 5% of patients undergoing proton therapy experienced hyperpigmentation.

3.2. Incidence of Dermatitis

In total, 18% experienced dermatitis while 82% did not. The individual who received both forms of radiation did not experience dermatitis. Fisher’s Exact Test was used to determine if there was a relationship between therapy type and incidence of dermatitis. The results revealed no significant association, p = 1.00. Due to assumption violations in the chi-square test of association, the likelihood ratio test of association was used to determine if there was a relationship between the RTOG score and therapy type. The results revealed no significant association, x = 1.32, p = 0.52. The Both group was excluded from both analyses due to a small sample size. Figure 1 and Figure 2 provides a full description of dermatitis and RTOG between groups.

3.3. Chemotherapy vs. Surgery

Figure 3 examines the association of dermatitis between patients who received chemotherapy, surgery, or both. The likelihood ratio statistic was used to determine if there was an association between those who received chemotherapy, surgery, or both. The results revealed a significant association between the groups and dermatitis, x2 = 6.25, p = 0.044. Those who received chemotherapy were more likely to have dermatitis than those who received surgery or both. The none group was not included in data analysis due to a small sample size.

3.4. Relationship Between Total Dose, Fraction Dose, and RTOG

The relationship between total dose, fraction dose, and RTOG can be expressed as follows: Individuals who experienced RID with an RTOG score of 1 had a median dose of 5040.0 (3240.0–5780.0) mGy, those with a score of 2 had a median total dose of 5020.0 (3825.0–6245.0) mGy and those with a RTOG score of 3 had a median total dose of 7600.0 (6660.0–7600.0) mGy. Those with a RTOG score of 1 had a median of 28 (21.3–31.5) fraction, those with an RTOG score of 2 also had a median fraction of 28 (21.3–34.8), and those with a score of 3 had a median fraction number of 30 (20.0–30.0). A nonparametric Kruskal–Wallis Test was used to determine if there was a significant difference between RTOG score and total dose. The results revealed significant differences among the groups, x2 = 6.4, p = 0.04. Dunn post-hoc tests revealed that those with an RTOG score of 3 had a significantly higher dose than those with a RTOG score of 1 (p = 0.04). There were no significant differences in fractions between the groups, x2 = 0.52, p = 0.77.

3.5. Time of Resolution

From the 15 individuals who had a time of resolution of their symptoms, 11 of those individuals received interventions with a median resolution time of 2 (1.0–3.3) months. Of the 4 who did not receive intervention, their median resolution time was 1.5 (0.8–2.7) months. Resolution of symptoms was defined as either achieving an RTOG score of 0 or a significant reduction in pain and discomfort, as reported by the patient. A Mann–Whitney U test was used to compare resolution time and no differences were found between groups, z = −0.39, p = 0.75. Of the 15 with documented resolution time, all of them received proton therapy with a median resolution time of 1.8 (1.0–3.0) months.

4. Discussion

This study aimed to determine the incidence and secondary characteristics of radiation-induced dermatitis among pediatric patients who received proton and/or photon therapy. The findings from this study offer insights into incidence, risk factors, and relationships in the severity of dermatitis in pediatric populations. These findings can contribute to a broader understanding on how different radiation modalities affect skin toxicity in the pediatric population.

4.1. Prevalence and Severity of Radiation-Induced Dermatitis

The results show that RID was a relatively common side effect, with 18% of the proton group and 15% of the photon group experiencing a form of dermatitis post radiation exposure. These findings are lower than previous reports stating that dermatitis can be seen in up to 95% of those receiving radiation therapy [13]. The difference in the rates in this study may be due to differences in patient population, treatment protocol, assessment methods, and challenges in evaluating skin toxicity through a standardized approach [14].

While the overall incidence of RID was similar across both proton and photon therapy groups (p = 1.0), the severity differed. In the proton group, a wider range of RTOG scores were seen with 71% of the patients having mild dermatitis (RTOG 1), 17% having moderate dermatitis (RTOG 2), and 12% having severe dermatitis (RTOG 3). In the photon group, however, all cases of dermatitis were mild (RTOG 1), but these differences were not significant. This difference in lack of severity scaling may be due to the limited sample size of the photon group. Further studies with a larger population should be used to determine if these results are consistent. It could also be from differences in delivery methods of radiation. Proton therapy has a more localized dosing method which could potentially lead to higher radiation in certain areas and thus more severe skin reactions [15].

4.2. Secondary Characteristics of RID

In terms of hyperpigmentation, 5% had evidence of discoloration within the proton therapy group and while there was none in the photon therapy group. These results seem to be substantially smaller than those seen in previous studies, which have shown a hyperpigmentation rate closer to 80% post radiation [16]. Differences could be from self-reported versus clinician-reported hyperpigmentation, time at which the data were collected, and population type.

Prior dermatological conditions were also documented due to research stating that a history of skin conditions can influence RID [17]. However, limited data were procured regarding past skin conditions, so analysis between severity and past history could not be conducted. Further research with a larger population size can be performed to look at this factor.

There was a greater number of individuals with acute toxicity (13%) versus late toxicity (8%). Studies have varied on whether there are differences between the amount of acute vs. late toxicity in radiation therapy.

Only 11% (10% in the proton group and 15% in the photon group) required a form of treatment for their dermatitis. Treatments involved corticosteroids, silver sulfadiazine, and pain relievers. A subset of the sample had time of resolution documented, and there were no significant differences in those who received intervention versus those who did not (p = 0.75). The time of resolution was similar to previous studies with dermatitis peak at 2 weeks and lasting from 1 to 2 months [15,17].

4.3. Risk Factors and Correlations

Multiple risk factors of RID were found, including total radiation dose and chemotherapy as an additional oncological intervention. A significant difference was found between the radiation dose and RTOG score, with higher doses of radiation associated with higher RTOG scores and more severe cases of dermatitis (p = 0.04). This finding is consistent with previous studies that have shown a dose-dependent relationship between radiation exposure and severity of RID [18]. There was no significant difference between fraction dose and RTOG score (p = 0.77). The lack of a significant difference between fraction dose and RTOG score suggests that dose, rather than the fractionation schedule, may be the primary driver of skin toxicity in this population. Studies have shown a direct correlation with fractionation in RID. However, it seems that with smaller fractions there is a higher level of radiation per treatment session; therefore, the severity of dermatitis may increase. Ultimately, the RID is still dependent on dosing [19]. The median total dose (5220.0 mGy) was consistent with previous studies showing pediatric focal radiation dosage at a median of 50.4 Gy, specifically in primary brain tumors [20].

The relationship between chemotherapy and RID was significant, with 36% of patients who had chemotherapy alone developing RID, compared to the 4% who only had surgery (p = 0.04). This finding supports the idea that chemotherapy can exacerbate skin toxicity, possibly by impairing the skin’s ability to repair damage caused by radiation [17]. However, it is important to consider the small sample size of patients receiving chemotherapy or surgery alone, further stating the need for investigation with larger samples.

4.4. Implications for Clinical Practice

The findings in this study have important implications for clinical practice. Firstly, the similar incidence of RID between proton and photon therapy suggest that both forms of therapy have comparable risks of skin toxicity within the population. Also, the significant differences between total radiation dose and dermatitis severity highlight the importance of optimizing radiation total dosing and dose per fraction to lower the risk of RID. This is particularly important in children, who are more susceptible to the long-term effects of exposure [4].

Additionally, the relationship between chemotherapy and higher rates of RID suggests that patients who receive both radiation and chemotherapy should be carefully monitored to prevent and appropriately treat dermatitis and lower the risk of long-term effects. Common treatments used include topical corticosteroids, emollients, and dressings to help reduce symptoms [21].

4.5. Limitations and Areas for Improvement

While this study provides insights into the incidence and risk factors of RID for both proton and photon therapy in the pediatric population, its limitations should be acknowledged. The retrospective study could introduce potential biases due to incomplete data. Additionally, the small sample size within the subgroups limit the generalizability of the results. The results with the small subgroups should be interpreted with caution due to the likelihood of reduced power of analysis. Another limitation was the exclusion of the dosimetry plan which would have given important information for the treatment plan.

To address these limitations in the future, studies should aim to include larger and more diverse patient populations to improve the power and generalizability of the findings. Prospective studies with standardized methods to assess RID and consistent monitoring would help reduce the variability in the results and allow for a more detailed comparison between the two therapy types. It would also be beneficial to record chemotherapy agents used simultaneously with radiation to determine if some agents are more prone to causing RID. Finally, investigating the long-term effects of radiation and monitoring potential chronic or late dermatitis within the pediatric population would provide valuable information into the life-long effects of radiation therapy on skin toxicity and quality of life [11].

5. Conclusions

This study provides information on the incidence and severity of RID among pediatric cancer patients treated with proton and photon therapies. The findings suggest that both radiation modalities have similar risks of causing skin toxicity, with no significant difference in the incidence of dermatitis between the two groups. However, the study shows that higher radiation doses were associated with more severe dermatitis, highlighting the importance of optimizing radiation dosing to minimize side effects, particularly in pediatric patients who are more likely to have long-term radiation toxicity. Additionally, the association between chemotherapy and the increased rates of dermatitis emphasizes the need for the monitoring of patients receiving combined treatments.

Despite this new information, this study also emphasizes the need for more research. The small sample size and retrospective nature of this paper limit the generalizability of the findings, and the absence of dosimetry data leaves gaps in the knowledge regarding the other factors impacting RID. Future studies with larger, more diverse populations, prospective data collection, and standardized RID assessment methods will help strengthen the evidence base. Understanding the long-term effects of radiation on the skin in pediatric populations remains critical for improving quality of life for cancer patients.