Abstract
Background/Aim: A significant number of breast cancer patients assigned to adjuvant radiotherapy receive pre- or postoperative chemotherapy that can cause peripheral sensory and autonomic neuropathy (PNP). Identifying and grading PNP using a mobile application (app) may facilitate early detection and management. This study evaluated the prevalence of pre-radiotherapy PNP, providing essential baseline data for the prospective validation of such an app.
Patients and Methods: Data of 133 breast cancer patients receiving chemotherapy prior to adjuvant radiotherapy were retrospectively analyzed regarding the incidence and risk factors of chemotherapy-induced moderate to severe PNP.
Results: The incidence of moderate to severe PNP was 27.8%. On multivariate analysis, it was significantly associated with KPS ≤80 (p=0.030), history of significant cardiovascular disease (p=0.024), ≥10 pack years of smoking (p=0.023), and beta blocker medication (p=0.014). A trend was found for history of autoimmune disease (p=0.057).
Conclusion: A considerable proportion of breast cancer patients assigned for adjuvant radiotherapy experienced moderate to severe PNP, which radiation oncologists should be aware of. The risk factors identified in this study likely contribute to earlier detection of PNP and inform the development of a prospective trial to validate a mobile-based PNP assessment tool.
Introduction
Adjuvant radiotherapy is a standard procedure following breast-conserving surgery for non-metastatic breast cancer (1-3). Depending of the tumor stage and other risk factors, radiotherapy may be indicated also after mastectomy. A considerable number of these patients receive pre- or postoperative chemotherapy prior to their course of irradiation. These chemotherapy regimens generally include taxanes (paclitaxel or docetaxel), which are known to be associated with a considerable risk of causing peripheral sensory neuropathy (PNP) (4). Radiation oncologists should be aware of the possibility of this adverse event. In the Common Toxicity Criteria of Adverse Events (CTCAE) version 5.0, PNP is described as a disorder characterized by damage or dysfunction of the peripheral sensory nerves (5). Symptoms of moderate to severe PNP are characterized by limitations in instrumental activities of daily living and, in severe cases, restrictions in self-care activities. Thus, moderate to severe PNP can be quite burdensome for the affected patients. We also have concerns that such distal PNP might also interfere with dermal integrity at the site of radiation through a more widespread autonomic neuropathy with reduced sudomotor activity.
According to three review articles, there is no effective prophylactic treatment to prevent PNP (4, 6, 7). Moreover, treatment options for existing PNP are very limited. The only agent shown to be effective to improve symptoms of PNP (i.e., pain) in a phase 3 trial is duloxetine, a serotonin-norepinephrine reuptake inhibitor (7-9). In addition, physiotherapy may contribute to maintaining or improving the patient’s gait function. Considering these limited options, it appears important to know risk factors of moderate to severe PNP. This knowledge may help identify patients who may benefit from more personalized chemotherapy regimens, Moreover, close monitoring during the treatment is important to be able to adjust the dose of neurotoxic chemotherapy very soon after the first symptoms of PNP occur, if reasonably possible. Depending on the progression and severity of PNP, adjustments to the chemotherapy regimen, including dose modification or discontinuation, may be necessary.
To prevent severe neurotoxicity during chemotherapy, the oncology service needs to follow early development of symptoms. In breast cancer patients treated with taxanes, such as paclitaxel, dose reduction due to early development of neuropathy does not change long-term outcomes of toxicity symptoms, suggesting that responses to first doses might be important (10). In addition, the patient and healthcare services burden limits frequent out-patient visits for early detection of toxicity. To document neurotoxicity screening according to CTCAE version 5.0 and the European Organisation for Research and Treatment of Cancer (EORTC), symptoms and their impact on function must be documented (5, 11). In clinical neurotoxicity studies, a great effort has been put into documenting the validity of grading scales, incorporating results from clinical examination findings with patient-reported symptom data (12). Both approaches require health care resources to document and follow up on PNP development in order to intervene in the case of severe early toxicity.
We developed Neuropath Tracker, a digital health mobile-based and patient-administered device for PNP grading based on patient reported symptoms and signs from self-examination (13). This solution for large scale data collection is currently under clinical evaluation for reliability and validity testing in Distal Symmetrical and Small Fiber Neuropathy. Patients at risk of experiencing symptomatic chemotherapy-induced PNP may be candidates for such a mobile application (app) used by the patients themselves at home. The present study that investigates PNP in breast cancer patients assigned to adjuvant radiotherapy is important for designing a prospective trial testing such an app as part of the German-Danish Interreg-project HeAT.
Patients and Methods
A total of 133 female breast cancer patients presented to the Department of Radiation Oncology at the University of Lübeck in the years 2022 or 2023 were included in this retrospective study, which received approval from the local ethics committee (file 2025-100). The patients were assigned to adjuvant radiotherapy following breast conserving surgery (N=102) or mastectomy (N=31). All patients received systemic chemotherapy prior to the start of radiotherapy, either before (neoadjuvant) or following (adjuvant) breast surgery. The chemotherapy regimens are summarized in Table I.
Details of the planned chemotherapy regimens.
Radiotherapy was performed as intensity-modulated radiation therapy (N=73) or volumetric modulated arc therapy (N=60). In the 127 patients with unilateral breast cancer, treatment volumes of radiotherapy included the whole breast alone in 65 patients, the whole breast plus regional lymph nodes in 35 patients, the chest wall alone in two patients, and the chest wall plus regional lymph nodes in 25 patients. Dose-fractionation regimens included hypo-fractionation with 40 Gy (15×2.666 Gy over three weeks) in 70 of the 127 patients or normofractionation with 50.4 Gy (28×1.8 Gy over 5.5 weeks) in 57 patients. In 99 of 127 patients, radiotherapy was supplemented by a boost to the former primary tumor site, either sequentially (5×2.0 Gy) or as a simultaneous integrated boost (SIB) of 0.3 Gy per fraction. Six patients had bilateral breast cancer. Two of these patients received unilateral irradiation, one to the chest wall alone (40 Gy) and one to the chest wall plus regional lymph nodes (50.4 Gy) without a boost. Of the other four patients, two each received 40 Gy or 50.4 Gy; treatment volumes and boost concepts varied depending on the affected breast and individual tumor characteristics.
Main endpoints of this study included the incidence of chemotherapy-induced moderate to severe PNP and identification of corresponding risk factors. Investigated potential risk factors included time of chemotherapy (neoadjuvant vs. adjuvant), type of surgery (breast-conserving surgery vs. mastectomy), chemotherapy regimen [epirubicin/cyclophosphamide (EC) + paclitaxel vs. EC + paclitaxel/carboplatin vs. docetaxel/carboplatin/trastuzumab/pertuzumab vs. paclitaxel±carboplatin or ±trastuzumab vs. epirubicin/paclitaxel/cyclophosphamide vs. EC + paclitaxel/carboplatin + pembrolizumab (Table I)], age at chemotherapy (≤64 years vs. ≥65 years=elderly), Karnofsky performance score (KPS ≤80 vs. 90-100), body mass index (BMI <25.0 kg/m2 vs. 25.0-29.9 kg/m2 vs. ≥30.0 kg/m2), personal status (living with spouse or partner vs. single/widow/living alone), history of another malignancy (no vs. yes), history of autoimmune disease (no vs. yes), history of significant cardiovascular disease (no vs. yes), hypertension (no vs. yes), diabetes (no vs. yes), history of smoking (<10 pack years vs. ≥10 pack years), beta blocker medication (no vs. yes), main histology (no special type alone vs. other), histologic grading (G1 or G2 vs. G3), and triple negativity (no vs. yes). The distribution of these characteristics is shown in Table II. Since both patients receiving neoadjuvant chemotherapy and patients treated with adjuvant chemotherapy were included, primary tumor stage and nodal stage were not uniform. Stages at the time of surgery could be pathological (pT and pN) or pathological following neoadjuvant chemotherapy (ypT and ypN). Moreover, clinical stages (cT and cN) were not always available. Considering these limitations that may have led to selection biases, it was decided to not include primary tumor stage and nodal stage in the analyses of risk factors for moderate to severe PNP. In patients with synchronous bilateral cancer, the more advanced tumor was considered. Other malignancies included contralateral breast cancer in six patients, malignant melanoma in two patients, cervix cancer in one patient, and thyroid cancer in one patient. Autoimmune diseases included Hashimoto’s thyroiditis in seven patients, rheumatoid arthritis in five patients, bronchial asthma in two patients, psoriasis plus/minus arthritis in two patients, and other autoimmune diseases in six patients. Significant cardiovascular disease included heart failure, coronary heart disease, or cardiomyopathy in or heart failure in five patients, history of thromboembolic complications in three patients, and advanced vascular disease in three patients.
Distribution of the patient characteristics in the entire cohort (N=133).
For the univariate analyses regarding associations between chemotherapy-induced moderate to severe PNP and investigated patient characteristics (Table II), the Chi-square test and the Fisher’s exact test were used. For multivariate analysis, logistic regression models were fitted. A stepwise selection approach was used to create the final parsimonious model, which includes the most important factors. For the most parsimonious model that adequately explains the data, significance levels of 0.15 and 0.20, were set for variable inclusion and retention, respectively. The analyses were performed with SAS 9.4 (SAS Institute Inc, Cary, NC, USA). p-Values of less than 0.05 were regarded significant, and p-values of less than 0.10 were considered indicating a trend.
Results
In the entire cohort, 85 patients (63.9%, 95% confidence interval=55.8-72.1%) developed at least mild chemotherapy-induced PNP. Thirty-seven patients were identified with moderate to severe PNP, corresponding to an incidence of 27.2% (95% confidence interval=20.2-35.4%). On univariate analyses, moderate to severe PNP was significantly associated with KPS ≤80 (p<0.001), body mass index ≥30.0 kg/m2 (p=0.008), history of autoimmune disease (p= 0.019), history of significant cardiovascular disease (p= 0.002), history of hypertension (p=0.034), ≥10 pack years of smoking (p=0.042), and beta blocker medication (p<0.001). A trend was observed for age ≥65 years at the start of chemotherapy (p=0.066). The results of all univariate analyses are shown in Table III. In the final parsimonious model after stepwise regression, KPS ≤80 (p=0.030), history of significant cardiovascular disease (p=0.024), ≥10 pack years of smoking (p=0.023), and beta blocker medication (p= 0.014) achieved significance (Table IV). In addition, a trend was found for history of autoimmune disease (p= 0.057).
Relationship between patient characteristics and moderate to severe peripheral sensory neuropathy (PNP) in the entire cohort (univariate analyses).
Multivariate analysis of moderate to severe peripheral sensory neuropathy (PNP) in the entire cohort.
Since the incidence of moderate to severe PNP was much lower in patients receiving TCbHP than in patients receiving other chemotherapy regimens (12.5 vs. 30.0-40.0%), additional analyses were performed in the 101 patients receiving other chemotherapy regimens than TCbHP (Table V). On univariate analyses of this subgroup (Table VI), moderate to severe PNP was significantly associated with KPS ≤80 (p=0.003), body mass index ≥30.0 kg/m2 (p=0.010), history of significant cardiovascular disease (p=0.005), and beta blocker medication (p<0.001). Trends were found for age ≥65 years at the start of chemotherapy (p=0.084), history of autoimmune disease (p= 0.084), and ≥10 pack years of smoking (p=0.098). In the final parsimonious model, ≥10 pack years of smoking (p= 0.025), and beta blocker medication (p=0.014) were significant; history of significant cardiovascular disease showed a trend (p=0.065) (Table VII).
Distribution of the patient characteristics in patients receiving other chemotherapy regimens than TCbHP (N=101).
Relationship between patient characteristics and moderate to severe peripheral sensory neuropathy (PNP) in patients receiving other chemotherapy regimens than TCbHP (univariate analyses).
Multivariate analysis of moderate to severe peripheral sensory neuropathy (PNP) in patients receiving other chemotherapy regimens than TCbHP.
Discussion
Many patients with breast cancer assigned to adjuvant irradiation following breast-conserving surgery or mastectomy receive chemotherapy with taxanes prior to the start of their radiotherapy course (2). One side effect of these chemotherapy regimens is PNP, which can significantly impair the patient’s activities of daily living (4, 5). In 19 previous studies, the incidence of chemotherapy-induced PNP in breast cancer patients ranged between 16% and 97% (10, 14-31). Considering this wide range, additional studies are required to better define the incidence of PNP. In the present study, the incidence of PNP of any level was 63.9%. This rate was similar to the rates of six previous studies (22-27). In these studies, the incidence of chemotherapy-induced PNP was 62%, 62.7%, 64.7%, 64.9%, 64.9%, and 65.9%, respectively. In 12 of the 19 previous studies, an incidence between 50% and 79% was reported (19-30). When considering all 19 studies, the median incidence was 62.7%, which was close to the incidence of 63.9% found in our study. Thus, the incidence of our study agrees with the findings of the majority of the previous studies investigating chemotherapy-induced PNP in breast cancer patients.
The present study mainly focused on moderate to severe PNP. The corresponding incidence was 27.8%. In the literature, we identified seven studies that reported the incidence of moderate to severe chemotherapy-induced PNP in breast cancer patients (18, 21, 23, 28, 31-33). In these studies, the incidence ranged between 10.4% and 34.6%. The incidence observed in our study was similar to four of the previous studies, which reported an incidence of 23.2%, 23.4%, 25.5%, and 27.7%, respectively (21, 23, 31, 32).
Unfortunately, treatment and prevention options for chemotherapy-induced PNP remain highly limited (4, 6-9). Therefore, it is important to identify patients with an increased risk of developing PNP during or following potentially neurotoxic chemotherapy prior to the start of treatment. These patients may benefit from close monitoring that facilitates early identification of PNP and adjustment of the chemotherapy regimen.
Therefore, another goal of this study was the identification of risk factors of moderate to severe PNP. According to our results, KPS ≤80, history of significant cardiovascular disease, ≥10 pack years of smoking, and beta blocker medication were significantly associated with moderate to severe PNP on multivariate analysis and, therefore, considered independent risk factors. In addition, history of autoimmune disease showed a trend. Moreover, BMI ≥30.0 kg/m2 and history of hypertension were associated with moderate to severe PNP on univariate analyses, and age ≥65 years showed a trend. Similar results were found in the subgroup analysis of patients receiving other chemotherapy regimens than TCbHP. In the corresponding multivariate analysis, history of significant cardiovascular disease, ≥10 pack years of smoking, and beta blocker medication were significant. KPS ≤80, BMI ≥25.0 kg/m2 were significantly associated with PNP on univariate analyses, and age ≥65 years and history of autoimmune disease showed a trend.
The prognostic role of some of these factors has been previously described. In most studies investigating potential risk factors for chemotherapy-induced PNP, a higher BMI was associated with an increased risk of this complication (15, 18, 21, 24, 30, 34-41). Seven of these studies included more than 200 patients (15, 21, 30, 34, 36, 38, 41). Song et al. compared a BMI ≥25.0 kg/m2 to <25.0 kg/m2 in a retrospective cohort of 1,516 breast cancer patients receiving docetaxel or paclitaxel (15). Taxane-induced PNP occurred in 25.8% (BMI ≥25.0 kg/m2) and 20.4% (<25.0 kg/m2) of the patients, respectively (p= 0.019 on multivariate analysis). Another study included 296 patients from a previous prospective study, who received taxane-based chemotherapy for early-stage breast cancer (21). The rates of chemotherapy-induced PNP were 48.4%, 60.2%, and 66.7% in patients with a BMI <25.0 kg/m2, 25.0-30.0 kg/m2, and >30.0 kg/m2, respectively (p=0.036). Lixian et al. investigated risk factors for chemotherapy-induced PNP in 350 breast cancer patients receiving treatment with paclitaxel (30). A BMI ≥24.0 kg/m2 was significantly associated with a higher incidence of PNP than a BMI <24.0 kg/m2 (87.4% vs. 69.4%, p<0.001). In the study of Greenlee et al. that included 1,237 breast cancer patients of the prospective Pathways Study treated with taxane-based chemotherapy, patients with a BMI ≥30.0 kg/m2 (odds ratio=3.21) and a BMI of 25.0-29.9 kg/m2 (odds ratio=2.37) had a significantly higher incidence of PNP when compared to patients with a BMI <25.0 kg/m2 (34). In an ancillary analysis of three prospective studies including a total of 298 patients who received treatment with taxanes, a BMI ≥30.0 kg/m2 was significantly associated with PNP (risk ratio=1.45, 95% confidence interval=1.01-2.08) (36). In a retrospective study of 400 patients receiving paclitaxel for early-stage breast cancer, PNP occurred in 40%, 46% and 55% of patients with a BMI of 18.5-25.0 kg/m2, 25.0-29.9 kg/m2, and ≥30.0 kg/m2, respectively (p=0.069) (38). In another study of 268 breast cancer patients treated with taxane-based chemotherapy, occurrence of moderate to severe PNP was significantly associated with a BMI ≥24.0 kg/m2 when compared to 22.0-24.0 kg/m2 and <22.0 kg/m2 (65.8% vs. 28.9% and 5.3%, p< 0.001) (41).
Seven studies reported older age to be a significant risk factor for PNP (15, 18, 21, 30, 39, 40, 42, 43). Four studies included more than 200 patients (15, 21, 30, 43). In the study of Song et al., rates of PNP were 11.5% in patients aged 20-40 years, 22.4% in patients aged 41-60 years, and 35.6% in patients aged >60 years (p<0.001), respectively (15). Bao et al. found a trend indicating that patients aged >65 years had a higher incidence of PNP when compared to patients aged ≤65 years (67.6% vs. 55.4%, p=0.066) (21). Lixian et al. compared the age groups 18-45, 46-59, and >60 years (30). PNP rates were 60.4%, 83.0%, and 94.5%, respectively (p<0.001). In the study by Sanchez-Barroso et al. that included 503 patients with breast cancer or ovarian cancer treated with paclitaxel, patients requiring modification of the paclitaxel dose were significantly older (median age 55.9 vs. 51.0 years) (43). Similarly, patients who developed grade 3 neurotoxicity were also older (median age 53.8 vs. 49.5 years).
Moreover, three studies found associations between an increased incidence of PNP and beta blocker medication (41, 43, 44), two studies between an increased incidence of PNP and history of hypertension (30, 40), and one study between an increased incidence of PNP and history of smoking (39). Other factors showing an association with moderate to severe PNP in our study, i.e., KPS ≤80, history of significant cardiovascular disease, and history of autoimmune disease, were not previously described in patients receiving taxane-based chemotherapy for breast cancer. However, Lee et al. mention in their review article that an overall burden of comorbidities may increase the risk of PNP (45). This burden often includes cardiovascular diseases, which can negatively impact the patient’s performance status (KPS). Moreover, cardiovascular disease can lead to a reduced blood flow and impaired oxygen supply of the peripheral nerves, which may aggravate PNP. In previous studies, we found that significant cardiovascular disease was a risk factor for radiation-induced toxicities, namely grade ≥2 dermatitis in breast cancer patients and grade ≥2 pneumonitis in lung cancer patients (46, 47). Autoimmune diseases such as Sjögren’s syndrome, rheumatoid arthritis, and systemic lupus erythematosus may lead to auto-antibodies that directly cause damage to the peripheral nerves (48). Recently it was reported that a subgroup of patients with autoimmune thyroiditis and hypothyroidism develop neuropathic symptoms (49). Moreover, autoimmune (chronic inflammatory) disease was found to be a risk factor for grade ≥2 dermatitis and pneumonitis in patients irradiated for breast cancer (46, 50). In contrast to the factors mentioned above, moderate to severe PNP was not significantly associated with diabetes in the present study, although the rate of PNP was much higher in diabetic patients than in patients without diabetes (50.0% vs. 26.4%). One may speculate that significance was not reached, because the number of diabetic patients was very small (N=8).
The fact that the incidence of moderate to severe PNP and the risk factors identified in our study are mainly in line with and can be explained by the results of previous studies supports the consistency and reliability of our data. However, the retrospective design of our study needs to be considered during the interpretation of its results. This design always bears the risk of hidden selection biases. Despite this limitation, the results of this study are important for the creation of a prospective trial. This trial will investigate the benefit of an app, which can be used by the patients at home and is expected to facilitate the detection of chemotherapy-induced PNP.
In summary, a considerable proportion of the patients treated with taxane-based chemotherapy prior to adjuvant radiotherapy for breast cancer experienced moderate to severe PNP, which radiation oncologists should be aware of. This study identified several risk factors including new predictors of moderate to severe PNP. Patients at a higher risk of developing PNP will likely benefit from closer monitoring during and following their treatment. Moreover, the results of this study are important for designing a prospective trial that will evaluate the effectiveness of an app for early detection of PNP.
Acknowledgements
As part of the Interreg-project HeAT, the study received partial funding from the European Regional Development Fund through the Interreg Deutschland-Danmark program (reference 01-1-23 2).
Footnotes
Authors’ Contributions
D.R., T.B., A.R., N.Y.Y., and M.B. participated in the design of the study. D.R. collected the data that were analyzed by a professional statistician. D.R. and M.B. drafted the article, which was reviewed and approved by all Authors.
Conflicts of Interest
The Authors indicate no conflicts of interest related to this study.
- Received March 11, 2025.
- Revision received March 20, 2025.
- Accepted March 21, 2025.
- Copyright © 2025 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
