Abstract
Background/Aim: Chronic secondary pain from cancer may flare into unexpected acute phases, called breakthrough cancer pain (BTcP). This phenomenon has a moderate-to-severe intensity and short latency between onset and peak of intensity, so clinical and therapeutic implications may be necessary. The Italian Oncologic Pain Multisetting-Multicentric Survey (IOPS-MS) aimed to characterize BTcP in a large number of patients from different settings and assess possible factors influencing its development. Previously, we observed differences according to sex in site, onset and level of BTcP. In this analysis, we assessed differences in non-predictable BTcP between female patients with female-specific cancer (FSC) (including breast and gynecological cancer) and those with non-FSC.
Patients and Methods: A univariate analysis compared patients with FSC (overall and separately for breast and gynecological cancer) and non-FSC. A multivariate analysis stratified by BTcP phenotype was performed to estimate the main determinants of the risk groups in patients with FSC (overall and separately for breast and gynecological cancer) and non-FSC. Odds ratios and 95% confidence intervals were reported.
Results: The FSC group was younger, more often treated in clinics or day hospitals, and had an increased risk of locoregional/metastatic disease than the non-FSC group. Patients with breast cancer were at higher risk of locoregional/metastatic disease, less frequently treated by a palliative specialist and had later BTcP onset compared to the non-FSC group. Significant differences concerning age, care setting and BTcP onset were also found in the group of patients with FSC (overall and separately for breast and gynecological cancer) according to tumor extension.
Conclusion: Non-predictable BTcP in patients with FSC presents unique characteristics, particularly regarding pain onset, care settings, and metastasis. Differences between breast and gynecological cancer emphasize the need for tailored pain-management approaches.
Introduction
In recent years, cancer mortality rates have declined in Italy and Europe. Between 2006 and 2021, a 21.4% reduction in cancer-related deaths was observed among women in Italy. Moreover, between 2007 and 2019, approximately 268,471 cancer deaths were avoided compared to the number expected in the period 2003-2006, highlighting the significant progress made in the field of oncology (1). Several key factors have contributed to this positive trend. Advances in diagnosis and treatment, including the implementation of screening programs and the development of more effective therapies, have played a crucial role in enabling early detection and improving patient outcomes.
Although cancer mortality has declined (1), most modern therapies focus on progress in cancer patients’ quality of life (QoL): they increase their lifetime, and the disease becomes chronic (2, 3). Pain is one of the worst symptoms that characterize a cancer patient’s clinical history. Indeed, approximately 30% to 85% of patients with cancer experienced chronic pain, depending on the type of cancer and the stage (4). The International Association for the Study of Pain distinguished pain into primary and secondary chronic pain types. Secondary chronic pain may explode through unexpected acute phases. This phenomenon is called breakthrough cancer pain (BTcP), a transient pain episode that occurs despite stable and well-controlled baseline pain in cancer patients. It has a rapid onset, reaching peak intensity within min, and typically lasts less than an hour (5). BTcP is classified into two categories, predictable and non-predictable BTcP (NP-BTcP) type. The predictable, or incident type, is further categorized into three sub-classes, including volitional BTcP (caused by a voluntary act), non-volitional (caused by an involuntary act), and procedural pain. NP-BTcP occurs in the absence of any specific activity. It is also indicated as spontaneous or idiopathic BTcP (6, 7).
The underlying pathophysiological mechanisms of BTcP are complex and multifactorial. A key factor is central sensitization, a process in which persistent stimulation makes the central nervous system hypersensitive to pain, leading to an exaggerated response even to mild inputs. BTcP can arise spontaneously or be triggered by specific factors such as movement or external inputs (8). Moreover, it may occur without prior indication, with a reported prevalence ranging from 40% to 80% among cancer patients experiencing pain, thus carrying significant clinical and therapeutic implications. In this regard, evidence indicates that poor management of BTcP severely affects clinical outcomes of patients with cancer, determining poor compliance with cancer treatments, increased healthcare costs, and patient dissatisfaction; moreover, it may negatively influence patients’ QoL, leading to poor functional ability and sleep patterns, increased anxiety and depression (9). The Italian Oncologic Pain Multisetting-Multicentric Survey (IOPS-MS) was launched to characterize BTcP in a large number of patients from different settings and to evaluate possible factors influencing its development. Further details about this study and previous analysis were described elsewhere (6, 7). Another critical concern regarding BTcP is how pain is experienced and perceived between the sexes. These differences are related to biological, psychological, and social factors, and can have important implications for the assessment and treatment of pain (10, 11). One key difference between men and women is related to the hormonal aspects. For example, estrogen and progesterone can affect pain sensitivity, with some studies suggesting that women may be more sensitive to pain during certain stages of the menstrual cycle. Similarly, testosterone may have analgesic effects in men, which can contribute to differences in pain sensitivity between the sexes (12). Social factors, such as gender roles and cultural expectations, may also influence the experience and expression of pain. For example, some studies suggest that men may feel pressure to appear stoic and tough in the face of pain, while women may be more likely to seek social support and express their pain more openly (4, 13). Psychological factors may also play a role in gender differences in pain. For example, some studies suggest that women may be more likely to emphasize pain or to have higher levels of anxiety and depression, which can exacerbate pain symptoms. On the other hand, men may be more likely to suppress or ignore pain, which may delay treatment and lead to worse outcomes as described by Bartely et al. (14).
The principal aim of this study was to characterize NP-BTcP in women with breast or gynecological cancer, representing two hormone-related, female-specific cancer types (FSC) compared with female patients with non-FSC.
Patients and Methods
Details from the IOPS-MS dataset. The IOPS-MS study was a multicenter survey that involved five palliative care units, seven oncology centers, and nine outpatient pain clinics. The study was proposed by an expert group of 27 Italian centers representative of different settings of cancer pain, and 21 centers agreed to participate. The primary aim of this study was to characterize BTcP in a large number of patients performed in different settings and to assess possible factors influencing its development. The secondary aim was to gather information about the diagnosis and management of BTcP and patient satisfaction with treatment (15).
The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board (PROTOCOLLO DI STUDIO IOPS-MS), Policlinico Tor Vergata (REGISTRO SPERIMENTAZIONI n.21/13), Roma, 20 February 2013. Each patient gave their informed consent to study participation.
The study inclusion criteria were the following:
- Age greater than 18 years;
- Diagnosis of cancer at any stage;
- Well-controlled and stable background pain with an intensity ≤4 on a 0-10 numerical rating scale (NRS);
- Presence of BTcP episode of moderate-to-severe intensity, clearly distinguished from background pain.
The exclusion criteria were:
- No cancer diagnosis;
- Unstable or uncontrolled background pain (NRS >4/10);
- No relevant peaks in pain intensity (NRS <5/10);
- Poor collaboration or refusal to participate.
Considering that in spontaneous BTcP, more than 3-4 episodes per day usually indicate uncontrolled background pain (requiring a careful optimization of basal pain) (16), we adopted the cut-off of 4 episodes per day. From the original study, some variables were selected for this secondary analysis: Age, setting, primary cancer type, tumor extension, BTcP onset time, BTcP pain type (nociceptive, neuropathic, or both), Karnofsky performance status score, and type of physician. Patients were asked about the average time of meaningful pain relief after their BTcP medication (classified as ≤10 min, >10 min). The study was observational, and pharmaceutical therapies were used according to local policy, without following strict protocols.
In our previous study, a multiple correspondence analysis and a hierarchical clustering principal-components analysis were adopted to interpret the BTcP phenomenon on the original IOPS-MS dataset that included 2,790 (69.6%) patients with NP-BTcP (6). The four clusters or phenotypes represent a classification of patients based on BTcP status, defined by variables of BTcP intensity, number of episodes and type. Phenotypes were classified from P1 to P4, from the best group (P1) to the worst (P4) and with the same features. Briefly, P1 was characterized by older age (≥75 years), slower BTcP onset (>10 min), gastrointestinal as primary tumor, and greater propensity to be treated in the context of the palliative care setting. On the contrary, the main features of P4 were: Younger age (<55 years) and rapid onset; furthermore, it most frequently concerned patients affected by lung cancer. Moreover, regarding the NP-BTcP therapy, P4 was mainly managed with rapid-onset opioids (ROOs); on the contrary, for P1, many patients were treated with oral, subcutaneous, or intravenous morphine. Moreover, the number of patients who did not receive therapy (ROOs, morphine, and other therapies) decreased from P1 to P4.
In the original IOPS-MS dataset, 69.5% (n=2,791) of participants had NP-BTcP. The statistical analysis of this study concerned the latter subgroup. Moreover, to investigate the differences in NP-BTcP among FSC (breast and gynecological cancers) and non-FSC groups within the subgroup of participants with NP-BTcP, only the subset of women was included in this secondary analysis (n=1,206).
Statistical analysis. A univariate analysis was performed to compare, within the population of patients with NP-BTcP, women with FSC with those with non-FSC. Statistical tests were used to compare means (Wilcoxon rank-sum test) and distributions (chi-square and Fisher exact tests) to determine the main differences between the two groups. A univariate analysis was also performed according to tumor extension (no metastases and locoregional/metastatic extension) to evaluate differences between FSC (overall and separately for breast and gynecological cancer subgroups) and non-FSC groups within the population with unpredictable pain. The four phenotypes (from P1 to P4) of NP-BTcP described elsewhere (6) were analyzed according to pain intensity, the number of episodes per/day, and the type of pain. A phenotype-like stratified multivariable analysis (by P1, P2, P3, P4) was performed to calculate the main determinants of the risk groups considered significant in the univariate analysis or with possible clinical interpretation. To identify the best associations, each model was evaluated stepwise. The odds ratios (ORs) and 95% confidence intervals (95% CIs) for significance are reported.
Results
The sample of women included in this analysis consisted of 1,206 patients: 813 (67.4%) had non-FSC while 393 (32.6%) had FSC, with 273 (22.6%) receiving a diagnosis of breast cancer and the remaining 120 (10.0%) a diagnosis of gynecological cancer.
Main characteristics of chronic pain according to tumor type. In Table I, the main characteristics of chronic pain for the FSC and non-FSC groups are reported. The FSC group was generally younger (mean of 61 vs. 64 years, p<0.001), enrolled more frequently in therapy of a Pain Department (33% vs. 25%, p<0.001), predominantly treated in a clinic or day hospital (53% vs. 41%, p<0.001) and less often in hospice or at home (10% vs. 20%, p<0.001). Moreover, the FSC group was more likely to have locoregional/metastatic cancer (93% vs. 82%, p<0.001) and to have later BcTP onset (>10 min: 38% vs. 29%, p=0.003).
Comparison of chronic pain characteristics between patients with female-specific cancer (FSC) and those with non-FSC.
Differences in BTcP characteristics between FSC and non-FSC groups according to tumor extension. In Table II, we analyzed the plausible differences in FSC and non-FSC groups in terms of BTcP characteristics according to tumor extension (no metastases and locoregional/metastatic), which might be a confounding factor in the group analysis. The group with locoregional/metastatic FSC was younger (mean of 61 vs. 64 years, p<0.001) compared to the non-FSC group. Moreover, for the locoregional/metastatic patient group, a difference in the care setting was confirmed: specifically, the FSC group was more frequently treated in a clinic or day hospital (52% vs. 37%, p<0.001) compared to the non-FSC group. Regarding medical care, we found that the group with locoregional/metastatic FSC was more frequently treated by pain therapists (34.5% vs. 23%, p<0.001). Moreover, a statistically significant difference was found in terms of BTcP onset among patients with locoregional/metastatic cancer: BTcP onset was later for the FSC group compared to the non-FSC group (>10 min: 38% vs. 28%, p<0.001).
Comparison of chronic pain characteristics in patients with female-specific cancer (FSC) and those with non-FSC according to tumor extension.
Differences in BTcP characteristics by FSC type. Patients with breast cancer: We further investigated the plausible difference between patients with breast cancer and the non-FSC group (Table III). Patients with breast cancer were younger (mean of 61 vs. 64 years, p<0.001), had a lower care setting in hospice or at home (7.7% vs. 20%, p<0.001) and were less frequently treated by palliative specialists (7.6% vs. 17%, p<0.001) compared to patients of the non-FCS group. Moreover, a higher proportion of patients with breast cancer had locoregional/metastatic disease (96% vs. 82%, p<0.001) and a delayed BTcP onset (>10 min: 38% vs. 29%, p=0.008).
Comparison of chronic pain characteristics in patients with female-specific cancer (FSC) and those with non-FSC.
Patients with gynecological cancer: In Table IV, we compared patients with gynecological cancer with the non-FSC group. Patients with gynecological cancer were generally younger (mean of 59 vs. 64 years, p<0.001) compared to the non-FSC group. No further statistically significant differences were observed in BTcP onset, treatment setting, or medical care.
Comparison of chronic pain characteristics in patients with gynecological cancer and those with non-female specific cancer (Non-FSC).
Differences according to tumor extension by FSC type. Patients with breast cancer: We then performed a univariate analysis by comparing patients with breast cancer and those of the non-FSC group according to tumor extension (no metastases and locoregional or metastatic cancer) (Table V). We found different behaviors, similar to what was reported in Table II. Patients with locoregional/metastatic breast cancer were younger (mean of 61 vs. 64 years, p=0.006) and had a higher care setting in a clinic/day hospital (55% vs. 37%, p<0.001). Moreover, they were less frequently treated by palliative care doctors (8% vs. 18%, p<0.001), more frequently treated by pain therapists (33% vs. 23%, p<0.001) and had a later BTcP onset (>10 min: 38% vs. 28%, p=0.003).
Comparison of chronic pain characteristics in patients with breast cancer and those with non-female specific cancer (Non-FSC) by tumor extension.
Patients with gynecological cancer: Table VI shows the results of univariate analysis comparing patients with gynecological cancer and the non-FSC group according to tumor extension. We found that patients with locoregional/metastatic gynecological cancer were younger compared to the non-FSC group (mean of 59 vs. 64 years, p<0.001). Moreover, gynecological cancer patients with locoregional/metastatic disease were less frequently treated by an oncologist (48% vs. 57%, p=0.032) and more frequently treated by a pain therapist (35% vs. 23%, p=0.032). Furthermore, in this group of patients with locoregional/metastatic disease, late BTcP onset was more frequent compared to the non-FSC group (40% vs. 28%, p=0.009). The comparison between the two types of FSC with the non-FSC group was further performed through a multivariate analysis (multivariate logit model). The analyses were stratified according to the four phenotypes (from P1 to P4) described above.
Comparison of chronic pain characteristics in patients with gynecological cancer and those with non-female specific cancer (non-FSC) group by tumor extension.
Multivariate analysis according to BTcP phenotype. FSC versus non-FSC groups: In Table VII, we compared the FSC and non-FSC groups according to the four phenotypes. In P1, we found that the FSC group had a three-fold increased risk of reaching a BTcP onset beyond 10 min (OR=3.01, 95% CI=1.46-6.22, p=0.003) compared to the non-FSC group. No statistically significant differences between the two groups were found regarding age. In P2, the FSC group was more likely to receive medical care from a pain therapist (OR=4.11, 95% CI=1.89-8.92; p=0.0001). Furthermore, for this phenotype, locoregional/metastatic cancer (OR=3.49, 95% CI=1.63-7.46; p=0.001) and age between 65-74 years (OR=0.46, 95% CI=0.258-0.819; p=0.008) were confirmed as characteristic factors of the FSC group. In P3, age was identified as a protective factor (OR<0.5 for each age level, p<0.05) for the FSC group, while locoregional/metastatic cancer remained a risk factor (OR=2.27, 95% CI=1.08-4.77; p=0.03) for the same group. In P4, age between 65-74 years (OR=0.31, 95% CI=0.132-0.755; p=0.01) was identified as a protective factor. Moreover, the FSC group had an increased risk of locoregional/metastatic cancer (OR=8.05, 95% CI=2.29-28.36; p=0.001).
Multivariable analysis comparing the female-specific cancer (FSC) and non-FSC groups by breakthrough cancer pain (BTcP) phenotype.
Multivariate analysis according to BTcP phenotype by FSC type. Patients with breast cancer: Table VIII shows the results of multivariate analysis comparing patients with breast cancer and the non-FSC group according to the four BTcP phenotypes. In P1, we found that patients with breast cancer had a two-fold increased risk of delayed BTcP onset compared to the non-FSC group (OR=2.46, 95% CI=1.12-5.41, p=0.025). Moreover, these patients also had a six-fold increased risk of locoregional/metastatic cancer, although this result did not reach statistical significance (OR=4.84, 95% CI=0.94-24.9; p=0.06); this risk remained high (OR greater than 6) for each phenotype. In P2, age represented a protective factor in patients with breast cancer aged between 65 and 74 years (OR=0.48, 95% CI=0.25-0.93; p=0.03). Moreover, we confirmed the association previously observed between breast cancer and increased risk of locoregional/metastatic cancer (OR=6.39, 95% CI=2.14-19.15; p=0.001) compared to the non-FSC group. Furthermore, patients with breast cancer were more likely to be treated by a pain therapist rather than a palliative care specialist (OR=5.57, 95% CI=2.12-14.66; p<0.0001) and had an increased risk of delayed BTcP onset (>10 min OR=1.65, 95% CI=0.997-2.73; p=0.05). In P3, patients with breast cancer had an increased risk of locoregional/metastatic disease (OR=5.93, 95% CI=1.71-20.5; p=0.005). In P4, age between 65-74 years was considered a protective factor, although this result did not reach statistical significance (OR=0.38, 95% CI=0.13-1.05; p=0.06).
Multivariable analysis comparing patients with breast cancer and those with non-female specific cancer by breakthrough cancer pain (BTcP) phenotype.
Patients with gynecological cancer: Finally, we performed a multivariate analysis comparing patients with gynecological cancer and the FSC group according to the four phenotypes (Table IX). For all the phenotypes, age between 65-74 years was identified as a protective factor particularly for P2, P3, and P4. In P1, patients with gynecological cancer had a 11-fold increased risk of delayed BTcP onset (>10 min: OR=11.7, 95% CI=1.99-68.2; p=0.006). The magnitude of this association decreased from P1 to P4, while maintaining the trend (for P4, OR=3.48, 95% CI=1.29-9.41; p=0.01).
Multivariable analysis comparing patients with gynecological cancer and those with non-female specific cancer by breakthrough cancer pain (BTcP) phenotype.
Discussion
NP-BTcP is the main topic of IOPS-MS (6, 7), the analyses being performed on this large dataset show ever more new pain characteristics and can potentially describe this phenomenon. In this secondary analysis, we aimed to investigate the plausible differences between breast and gynecological cancer, two hormone-related FSC, and the non-FSC group in terms of NP-BTcP, overall and across the four phenotypes. The latter defines the severity of BTcP and was calculated by combining BTcP intensity with the number of BTcP episodes per day and the BTcP type (nociceptive, neuropathic, or both) [see (6, 7)]. These phenotypes were characterized for the whole NP-BTcP sample in IOPS-MS, composed of male and female patients with cancer. We have shown that patients of the FSC group were younger, principally treated in clinics or day hospitals, and had a higher risk of locoregional-metastatic cancer compared to patients of the non-FSC group. Patients with breast cancer were at higher risk of locoregional/metastatic disease, more frequently treated by a pain therapist and more likely to have later BTcP onset compared to the non-FSC group. These findings confirm the data reported by Doan et al. (17). Specifically, the authors stated that the presence of metastases represented a significant risk factor for the onset and progression of pain in patients with breast cancer. Different studies showed that pain severity progressively increases with tumor size and involvement of adjacent structures. This progression often leads to tissue damage and an enhanced inflammatory response. Moreover, the localization of metastases in vital organs such as the brain, liver, and lungs can induce mass effects, resulting in nociceptive pain, in addition to painful localized inflammation, tissue destruction, and nerve injury (18-20). In literature, evidence indicates that women tend to have a higher occurrence of many clinical pain conditions, greater pain sensitivity, enhanced pain facilitation, and reduced pain inhibition compared with men, although the magnitude of these sex differences varies across studies (14, 21, 22). Moreover, hormonal replacement therapy use has been associated with an increased risk for back pain, temporomandibular disorders, carpal tunnel syndrome (4), and more severe pain (23). Regarding the effect of sex hormones on pain in patients with breast and ovarian cancer, evidence indicates that estrogen mediates pro-nociceptive actions through activation of the G-protein-coupled estrogen receptor, producing both spontaneous pain behaviors, as well as potentiating pain responses (24). We have shown here that patients with locoregional/metastatic gynecological cancer were younger, more frequently treated by pain therapists, and had later BTcP onset compared to the non-FSC group. In this regard it is important to underline that in patients with gynecological cancer, the development of chronic pain involves multiple mechanisms as described by Wu et al. (25). Specifically, ongoing inflammation and tissue damage can sensitize peripheral pain receptors, enhancing response and lowering the threshold for pain. In turn, the persistent stimulation of nociceptors can impair their excitability and induce the release of pro-inflammatory mediators, intensifying the pain response. Additionally, chronic pain can lead to neuroplastic changes in the central nervous system, making pain perception more intense and harder to manage (26). Moreover, psychological factors such as stress, anxiety, and depression play a role by heightening pain sensitivity and impairing patients’ QoL (27-29). Particularly as stated by Woo et aI. (30), symptoms of anxiety and depression should be taken into account in hospital and palliative care settings to ameliorate patients’ QoL.
Taken together, these findings highlight the need for further well-designed prospective studies with large populations aimed at thoroughly understanding the distinctive characteristics of BTcP, particularly NP-BTcP in women with FSC, to develop the most appropriate and personalized therapeutic strategies, improving patient outcomes and QoL.
Conclusion
BTcP is one of the most common and debilitating symptoms in patients with cancer. NP-BTcP among women with FSC presents unique characteristics, particularly regarding pain onset, care settings, and metastasis. Differences between breast and gynecological cancers emphasize the need for tailored pain-management approaches.
Acknowledgements
The Authors thank all study participants for their involvement in the study. The Authors also thank Maria Cristina Romano for data curation and the IOPS-MS group for their work in the original study. This work was (partially) supported by the Italian Ministry of Health 5X Mille funds, YEAR 2022.
Footnotes
Authors’ Contributions
A.C., S.B, and A.C. initiated and supervised the project. M.C., F.B., and M.M. collected the data. S.C., M.P., E.C., and A.L. performed the data analysis. S.B., A.C., D.S., D.N., A.A. interpreted the experimental data and prepared figures. A.C. and S.B. wrote the manuscript with input from all Authors. All Authors have been involved in the Manuscript’s revisions.
Data Sharing Statement
All data are available here and at the link DOI: 10.5281/zenodo.15654819
Conflicts of Interest
The Authors declare that there are no conflicts of interest in this study.
Funding
Molteni Farmaceutici, Italy, sponsored the original study. Data were independently analyzed and managed by the Authors. The funders had no role in the design of these secondary analyses or interpretation of data, in the writing of the manuscript, or in the decision to publish the results.
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received March 18, 2025.
- Revision received June 5, 2025.
- Accepted June 6, 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).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.