Sage Journals: Discover world-class research

Abstract

Background:

Radiotherapy (RT) brings a broad spectrum of side effects that could affect patient well-being. Pelvic insufficiency fractures (PIF) are one of them.

Objectives:

The aim of our study was to identify easily detectable risk factors for radiation-related PIF.

Design:

Prospective, single-center study.

Methods:

We included 104 patients aged 52.9 ± 13.8 years following radical RT for advanced cervical cancer. Patients underwent a pretreatment computed tomography (CT) imaging and a minimally 1-year follow-up by CT or magnetic resonance imaging. We evaluated the association between pretreatment CT attenuation values of L1, their deviation from normative values, age, body mass index, total received radiation dose, smoking habits, and radiation-related bone side effects.

Results:

In 28 (26.9%) patients PIF were found and first detected at a mean of 16 ± 7 months after RT. Patients with PIF were significantly older; 62.5 ± 10.2 versus 49.4 ± 12.6 years, p > 0.001; their pretreatment CT L1 attenuation values were significantly lower; 117.5 ± 46.9 HU versus 165.9 ± 44.8 HU, p < 0.001, as well as more negative deviation from normative values. Age and L1 attenuation values were strongly correlated, p < 0.001, precluding separation of their independent effects on PIF occurrence. According to logistic regression modeling, a 50-year-old woman had an estimated 16.3% probability of PIF (95% CI [8.6%; 25.9%]); the associated odds ratio increased by approximately 182% [72%; 357%] per 10-year increase in age. Thus, the estimated probability of PIF increased to 34.1% for a 60-year-old and 58.0% for a 70-year-old woman. The pretreatment CT attenuation values of 100 Hounsfield units (HU) were associated with a 51.1% probability of PIF (95% CI [36.2%; 66.5%]), and the probability decreased at higher attenuation values (odds ratio 0.766 [0.663; 0.865] per 10-HU increment). No other variables showed significant associations.

Conclusion:

Increasing age and lower pretreatment CT L1 attenuation values are strong predictors for radiation-related PIF, reflecting osteoporosis status.

Plain language summary

Bone health risk after radiation for cervical cancer.

Radiation for cervical cancer brings bone health risk as pelvic fractures. Women above 50 years old could be in risk due to osteoporosis. The paper brings data about possible selection of females in risk from the basic pretreatment examination which is CT scan. A simple measurement of lumbar vertebra bone densities is available to stratified females in risk.

Keywords

side effects bone densities predictor osteoporosis

Introduction

Cervical cancer is considered a nearly completely preventable cancer, and its incidence has been continually declining in developed countries during the past decades.¹ Despite this fact, cervical cancer remains the fourth most frequently diagnosed cancer in women worldwide, with an estimated 342,000 deaths annually.¹ Radiotherapy (RT) with concurrent chemotherapy is a standard treatment option for patients with locally advanced cervical cancer.² The combination of external beam radiation therapy (EBRT) and intracavitary brachytherapy (BRT) maximizes the locoregional control and minimizes the risk of radiation-related side effects.² Despite advances in modern RT techniques to reduce adverse events, some are difficult to avoid altogether. Radiation-related bone changes belong to these events. It is known that radiation alters several components of bone microstructure, causes vascular injury and devascularization, and probably directly impairs major bone cells such as osteoblasts, osteocytes, and osteoclasts. These cells are crucial for the bone matrix. Osteopenia after radiation is caused by reduced collagen formation and alkaline phosphatase activity.^3,4 In the radiation field, fat replacement of the bone marrow is commonly seen in imaging methods, especially on magnetic resonance imaging (MRI). Therefore, radiation-related insufficiency fractures could be expected, especially in the irradiated weight-bearing regions such as the pelvic bones or femoral heads.⁴ In patients treated with RT for gynecological cancers, pelvic insufficiency fractures (PIF) are relatively common radiation-related events that could affect the overall well-being of females. Despite the frequency of these side effects, some questions remain unanswered. The first is the accurate incidence of radiation-related bone side effects, which is still unknown. According to published data, the cumulative incidence of radiation-related PIF after radical RT for cervical cancer ranges widely from 8.2% to 45.2%.^4,5 This wide range may be partly because PIF are very often clinically overlooked; thus, their symptoms could be minimal or underestimated by patients or medical doctors.^4,5 The second question is which predictive factors can help identify patients at increased risk. Previous studies have suggested that age, body mass index (BMI), postmenopausal status, and various types of treatment are potential factors closely associated with PIF after RT.^4,5 The last question is whether preventive treatment could help prevent PIF in patients at risk.

Our team has been coping with radiation-related side effects after RT for gynecological cancers for years.^6–8 In our previous retrospective study, we evaluated available imaging scans in patients after RT for cervical and endometrial cancer.⁸ We anticipated that patients treated for endometrial cancer would have a higher incidence of PIF due to significantly higher age. However, the incidence of PIF was nearly identical in significantly younger females treated for cervical cancer, 28.6% versus 26.6%.⁸ Moreover, we also retrospectively measured attenuation values of the L1 vertebral body on pretreatment computed tomography (CT) and concluded that the radiation-related PIF rate was associated with lower bone densities in both groups.⁸

We hypothesized that radiation-related PIF could be predictable. Our present study was based on the results of our above-mentioned retrospective study. We hypothesized that the most important risk factor for PIF occurrence is the status of osteoporosis. According to the population-based study by Jang et al., osteoporosis status could be easily identified without any additional costs by measuring the CT attenuation values of L1 vertebral bodies. We followed the methodology of the above-mentioned study.⁹ The primary aim of the study was to measure attenuation values of the L1 vertebral bodies on pretreatment CT scans, compare them with normative ranges reported in the literature, and evaluate their association with PIF occurrence.⁹ The secondary aim was to assess the influence of age, BMI, total received RT dose, and smoking habits on PIF occurrence.

Materials and methods

The study was conducted as a prospective single-center study (Faculty Hospital Kralovske Vinohrady, Prague, Czech Republic). We included all consecutive patients with locally advanced cervical cancer who were treated with radical RT from January 2019 to February 2024. Patients treated between 2019 and 2021 were included as a secondary analysis from a previously published study cohort.⁸

The exclusion criteria were as follows:

Presence of bone fractures on pretreatment scans within the RT field;

Known trauma during the follow-up period;

Presence of bone metastases;

Death within the first year after RT;

Unwillingness or inability to undergo follow-up imaging (CT/MRI), or refusal to participate in the study.

At our institution, patients are routinely scheduled for an initial pretreatment MRI using a full diagnostic protocol as well as pretreatment non-contrast-enhanced CT scans for RT planning purposes. After RT, patients are usually scheduled for annual whole-body contrast-enhanced CT follow-up; this practice was followed in the present study. Additional MRI examinations 12 ± 2 months after completion of RT were offered to all patients. Moreover, some patients underwent additional, non-scheduled diagnostic MRI or CT examination according to the clinical need. All the available additional examinations were also evaluated, and data derived from these scans were included in the analysis.

We collected all relevant available clinical data and calculated the total received radiation dose as the sum of all administered doses (EBRT, BRT). BMI was calculated from the patient’s height and weight measured prior to RT. Smoking habits were recorded, and smoking exposure was quantified as pack-years for each smoker. A potential source of bias was the broad variability in adjuvant/neoadjuvant chemotherapy, as concomitant chemotherapy had to be suspended in some patients due to bone marrow suppression. Therefore, in each subject, we reported only the type of chemotherapy treatment (concomitant versus neo/adjuvant). The study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Ethics Committee of Faculty Hospital Kralovske Vinohrady, Prague, Czech Republic (EK-VP/22/00/2019; EK-VP/61/0/2020). All the patients provided written informed consent prior to inclusion. The study was reported in accordance with the STROBE cohort reporting guidelines.¹⁰

MRI and CT examinations, scan evaluations, and L1 attenuation values measurement

All the MRI examinations were performed as non-contrast-enhanced scans on 1.5T MR scanners (Magnetom Sola and Magnetom Altea; Siemens Healthineers, Erlangen, Germany). The MRI protocol covered the pelvis and consisted of the following sequences: TSE T2-weighted imaging (WI), slice thickness 3.5 mm, in three planes adjusted to individual patient anatomy; TSE T2 WI with fat saturation (FAT SAT), slice thickness 5 mm, in the coronal plane; TSE T1 WI, slice thickness 5 mm, in the axial plane; diffusion-weighted imaging with b-values of 50–800 s/mm², including apparent diffusion coefficient maps in the axial plane. Additional sequences included the following: T2 TRUFI FAT SAT, slice thickness 6 mm, in the coronal plane covering the pelvis and most of the abdominal cavity for assessment of lymph nodes, kidneys, and the urinary tract; TSE T2 WI FAT SAT, slice thickness 5 mm, in the coronal plane covering pelvic bones, hips, and lumbar spine for assessment of bone structures.

All the pretreatment CT scans were performed as non-contrast-enhanced examinations covering the abdomen and pelvis using a 128-slice CT scanner (Definition AS+; Siemens Healthineers). Follow-up CT examinations were performed using the same type of scanner as contrast-enhanced scans covering the chest, abdomen, and pelvis (“torso” scan, as a standard for European hospitals). CT acquisition parameters were as follows: rotation time 0.5 s, tube peak 120 kV, tube current modulation with quality reference value 140 mAs, iterative reconstruction iteration strength 3, soft tissue reconstruction filter/kernel, primary reconstruction slice thickness/increment 3/3 mm, and multiplanar reconstruction slice/increment 3/3 mm.

L1 attenuation values were measured on a pretreatment non-contrast-enhanced CT scan by placing a circular region of interest of 4 cm² in the center of L1 vertebral body. Attenuation values were recorder in Hounsfield units (HU); all measurements were performed on CT Bone Reading software (Syngo.Via, version VB60; Siemens Healthineers). For each patient, the measured L1 attenuation value was compared with normative L1 trabecular attenuation values published by Jang et al.⁹ The difference between actual measured and normative attenuation values was also included in the analysis.

On CT or MRI follow-up examinations after RT, scans were evaluated for radiation-related bone side effects such as PIF, vertebral compression fractures, and osteonecrosis within the radiation field. PIF was diagnosed only when fracture lines were clearly visible on T1 WI and T2 WI or CT images using a bone window. Isolated bone edema on T2 WI FAT SAT, without visible fracture lines, was classified only as radiation-related structural changes of bone marrow. Bone pathologies outside the radiation field were not considered radiation-related.

All the imaging examinations were independently evaluated by two board-certified radiologists.

Statistics

An experienced statistician performed all statistical tests and evaluations. We report descriptive statistics for the whole cohort of subjects. Group comparisons on continuous variables were examined using independent sample t-tests when variables displayed normal distribution (as determined by visual examination of histograms and Q-Q plots). Wilcoxon rank-sum tests were used instead for continuous variables with non-normal distributions. Categorical variables were compared using chi-square independence tests. Then, we used Bayesian logistic regression modeling to examine the probability of PIF based on L1 attenuation values and age. Bayesian logistic regression was chosen to better quantify the strength of evidence in favor of the hypotheses and to produce intuitive estimates (posterior probability distributions) for parameters. Posterior distributions were obtained by specifying weakly informative priors and using Markov Chain Monte Carlo methods implemented in the R package brms¹¹. Statistical significance was defined as p < 0.05 (two-tailed) for frequentist tests, and Bayesian results were summarized using posterior estimates and 95% credible intervals. Analyses were conducted in R version 4.4.3¹², with Bayesian logistic regression performed via the brms package, while frequentist comparisons (t-tests, chi-square, and Wilcoxon tests) were conducted using base R.¹³

Results

Patient selection

During the study period, 126 patients underwent radical RT for cervical cancer, of whom 36 patients were included as a secondary analysis from the previously published study cohort.⁸ However, 22 patients were not included in the study. Twenty patients refused participation or declined follow-up at our institution, one patient sustained a pelvic fracture in a car accident, and one patient died within the first year after RT. Finally, 104 patients aged 52.9 ± 13.8 years (median 54 years) were included in the study. The majority of included patients (N = 82) were treated for squamous cell carcinoma, while the remaining patients had adenocarcinoma or rare histological carcinoma subtypes. Patients were followed with imaging examinations for a minimum of 1 year and a maximum of 5 years. All the patients underwent MRI and/or CT scan 12 ± 2 months after completion of RT. The mean number of imaging examinations per patient, including additional scans performed due to clinical indications, was 5 ± 4. Additional scans were most commonly indicated for nonspecific complaints or suspicion of tumor recurrence. However, seven patients were examined for different pelvic fistulas, five patients for bowel perforation, and seven patients for hydronephrosis.

All the included patients underwent EBRT using intensity-modulated radiation therapy, most commonly with a dose of 45 Gray (Gy) delivered in 25 fractions. Eighty-seven patients underwent intracavitary BRT guided by 3D imaging, typically with a dose of 6.5 Gy delivered in four fractions. Some patients received additional RT to the parametria or lymph nodes with variable doses. Twenty-four patients underwent surgical treatment before RT or during the study period, mostly radical hysterectomy and adnexectomy for primary disease or surgical revision for early pelvic recurrence. Eighty-five patients received concomitant chemotherapy with cisplatin, administered once weekly for five cycles starting on day 1, at a dose of 40 mg/m² (maximum absolute dose of 70 mg). Eight patients received adjuvant or neoadjuvant chemotherapy. Eleven patients did not receive chemotherapy.

Bone findings on the follow-up imaging

Radiation-related bone side effects were identified in 28 patients (26.9%). In the vast majority of cases (N = 27), the sacrum was affected. In six cases, sacral PIF was combined with other radiation-related pathology: three with osteonecrosis of the pubic bones, one with osteonecrosis of the femoral head, and two with a combination of vertebral compression fracture. The first PIF was detected at a mean of 16 ± 7 months after completion of RT. Seventeen patients (60.7%) did not actively report any subjective symptoms. Nine patients reported nonspecific complaints, most commonly pain. In only two patients (7.1%) intervention and hospitalization were required; one of these patients underwent surgical treatment.

Analysis of subgroups with and without radiation-related insufficiency fracture

The included patients were divided into two subgroups: those with and those without radiation-related PIF. The subgroups were compared with respect to the following variables: pretreatment L1 attenuation values and their deviation from the normative values,⁹ age, BMI, total RT dose, type of chemotherapy, use of BRT, and smoking habits. Complete data are summarized in Table 1.

Table 1.

Characterization of included subjects and cohorts.

Variables	The whole cohort	The cohort with PIF	The cohort without PIF	p value
Number of subjects	104	28	76
Age (mean ± SD; median; min.–max.) years	52.9 ± 13.3; 54; 25–77	62.5 ± 10.2; 63; 42–77	49.4 ± 12.6; 49.5; 25–75	<0.001
Pretreatment bone densities of L1 (mean ± SD; median; min.–max.) HU	152.9 ± 50.0; 141.5; 44–273	117.5 ± 46.9; 114.5; 44–250	165.9 ± 44.8; 170.0; 85–273	<0.001
The difference in L1 attenuation values from normative values⁹ (mean ± SD; median) HU	−24.7 ± 25.8; −24.5	−35.5 ± 27.5; −34	−20.7 ± 24.9; −23	<0.019
Total received RT dose (mean ± SD; median; min.–max.) Gy	73.2 ± 14.7; 73.0; 45–128	73.3 ± 8.4; 73.0; 45–81	73.2 ± 16.4; 73.8; 45–128	0.973
EBRT received dose (mean ± SD; median; min.–max.) Gy	51.6 ± 8.8; 54.5; 45–100	50.2 ± 10.0; 50; 45–61.5	52.1 ± 10.0; 55; 45–100	0.504
BMI (mean ± SD; median; min.–max.) kg/m²	25.9 ± 6.2; 24.2; 16–49.4	25.6 ± 5.5; 24.2; 17.4–40.2	26.0 ± 6.0; 24.5; 16–49.4	0.721
BRT (yes)	87 (83.7%)	26 (92.9%)	61 (80.3%)	0.123
Concomitant chemotherapy (yes)	85 (81.7%)	23 (82.1%)	62 (81.6%)	0.826
Smoking (yes)	52 (50%)	17 (60.7%)	35 (46.1%)	0.269
Smoking (mean ± SD; median; min.–max.) years-pack	17.1 ± 11.6; 15; 0.5–50	20.3 ± 13.2; 20; 5.5–50	15.6 ± 10.7; 15; 0.5–41.3	0.208

PIF: pelvic insufficiency fractures; HU: Hounsfield units; RT: radiotherapy; Gy: Gray; BMI: body mass index; BRT: intracavitary brachytherapy; EBRT: external beam radiation therapy; SD: standard deviation.

Statistically significant differences were found only in age and pretreatment L1 attenuation values, as well as in their deviation from normative values published by Jang et al.⁹ Patients with radiation-related side effects were significantly older (62.5 ± 10.2 years versus 49.4 ± 12.6 years; t(59.30) = 5.44, p < 0.001, d = 1.09). Pretreatment L1 attenuation values were significantly lower in patients with bone side effects (117.5 ± 46.9 HU versus 165.9 ± 44.8 HU; t(46.32) = 4.72, p < 0.001, d = 1.07). It was not possible to statistically separate the influence of age and pretreatment L1 attenuation values on the presence of PIF because both variables were strongly correlated (r(102) = −0.75, p < 0.001). As shown in Figure 1, only four patients with PIF were younger than 50 years. Only seven patients had pretreatment attenuation values of L1 greater than 125 HU (five of seven had attenuation values exceeding 150 HU). L1 attenuation values in the entire cohort were lower than those reported in the published data by Jang et al.⁹ Patients with PIF had significantly larger negative deviations from normative L1 attenuation values, p = 0.019; see also Table 1. Measured L1 attenuation values were strongly correlated with normative attenuation values (r(102) = 0.73, p < 0.001; Figure 2).

Figure 1.

Association between age and pretreatment L1 attenuation values in relation to radiation-related fractures.

Figure 2.

Association between normative attenuation values and pretreatment L1 attenuation values in relation to radiation-related fractures.

To further illustrate the effects of age and pretreatment L1 CT attenuation values, we modeled the probability of the radiation-related side effects (see Figures 3 and 4). The model indicated that a 50-year-old woman has a 16.3% probability of radiation-related side effects (95% CI [8.6%; 25.9%]), with the associated odds ratio increasing by approximately 182% [72%; 357%] for each additional 10 years of age. In other words, the estimated probability of radiation-related side effects increased to 34.1% [23.7%; 45.8%] for a 60-year-old woman and to 58.0% [40.6%; 74.7%] for a 70-year-old woman. Similarly, pretreatment L1 attenuation values of 100 HU were associated with a 51.1% probability of PIF (95% CI [36.2%; 66.5%]), with the probability decreasing as attenuation values increased (odds ratio 0.766 [0.663; 0.865] per 10-HU increment).

Figure 3.

Probability of radiation-related fractures in relation to pretreatment attenuation values of L1.

Figure 4.

Probability of radiation-related fractures in relation to age.

Discussion

In our prospective study, the rate of radiation-related bone side effects after radical RT for cervical cancer was 26.9%. Sacral PIFs were identified in most cases and were first detected at a mean of 16 ± 7 months after completion of RT. We evaluated the association between PIF occurrence and pretreatment CT L1 attenuation values, including their deviation from normative values published in the literature,⁹ as well as other variables such as age, BMI, total RT dose, type of chemotherapy, BRT, and smoking habits. Between the cohort of patients with and without PIF, statistically significant differences were observed only for the pretreatment CT L1 attenuation values and age.

We found a strong, significant association between pretreatment L1 attenuation values and the PIF occurrence (117.5 ± 46.9 HU versus 165.9 ± 44.8 HU). Differences between measured L1 attenuation values and age-adjusted normative values were significantly more negative in patients with PIF. The normative CT L1 attenuation values published by Jang et al. were considered appropriate for comparison, as they were derived from a large cohort of 20 274 adult subjects spanning the full adult age range.⁹ These findings are consistent with reduced bone mineral density and suggest underlying osteoporosis.

Our results are in agreement with the study of Pickhardt et al.¹⁴ They compared the CT L1 attenuation value with bone mineral density measured by dual-energy X-ray absorptiometry in 1867 adult subjects. They reported that an L1 attenuation threshold of ⩽160 HU was 90% sensitive for osteoporosis, while a threshold of ⩽110 HU was more than 90% specific for osteoporosis.¹⁴ In our study, only five subjects with PIF had L1 attenuation values above 150 HU, and Bayesian modeling indicated that a pretreatment attenuation value of 100 HU was associated with a 51.1% probability of PIF, with the probability decreasing with each 10-HU increase in attenuation. Moreover, Salcedo et al., in their prospective study, demonstrated acceleration of osteoporosis development following RT for gynecological malignancies.¹⁵ They reported the increasing proportion of subjects with osteoporosis from 50% at baseline to 59% 1 year and to 70% 2 years after RT.¹⁵ In the general population, decline in bone mineral density is typically linear and substantially slower, with a reported mean decrease in CT L1 attenuation of 2.5 HU per year.⁹ Several additional studies showed that pretreatment densities on CT scans were significantly lower than in the subjects without PIF.^8,16 However, these studies were retrospective and subject to multiple sources of bias. In a study group by Kurrumeli et al.,¹⁶ they included 62 subjects with cervical cancer, and PIF was found only in 6 patients. In our previous retrospective study, we included 127 subjects, one half treated for cervical cancer and the other half for endometrial cancer, of whom 33 suffered from radiation-related PIF. However, the subgroups differed significantly with respect to age, BMI, RT dose, and chemotherapy regimens. Moreover, the follow-up imaging of included patients was irregular, performed using heterogeneous protocols and image quality, and a substantial number of patients were excluded due to missing imaging scans.⁸

In the present study, pretreatment CT L1 attenuation values strongly correlated with age. Therefore, it was impossible to calculate the influence of each variable on PIF occurrence. Age itself was strongly associated with PIF occurrence; patients with PIF were significantly older, 62.5 ± 10.2 years versus 49.4 ± 12.6 years. Bayesian logistic regression indicated that the estimated probability of PIF for a 50-year-old woman was 16.3% and then increased to 34.1% for a 60-year-old woman and 58.0% for a 70-year-old woman. The association between radiation-related bone side effects and increasing age or postmenopausal status has been repeatedly reported in previous studies.^5,8,17 The relationship between age and bone mineral density is also well established and was quantified by Jang et al. using CT L1 attenuation values.⁹ They reported that women, until menopause, have higher mean L1 attenuation value than men; however, after menopause, women and men have similar attenuation values with a mean decrease of 2.5 HU per year.⁹

Contrary to expectations, we did not observe significant differences in smoking habits between patients with and without PIF. According to the literature, tobacco smoking causes an imbalance in bone turnover, impairs osteogenesis and angiogenesis, and, therefore, is a risk factor for osteoporosis.¹⁸

Low BMI is considered another factor for radiation-related PIF development, as was reported in previous studies.^17,19 Previous reports have suggested that lower BMI may be associated with reduced mechanical loading and lower circulating estrogen levels, both of which are linked to decreased bone mineral density and increased osteoporosis risk. However, neither our prospective nor earlier retrospective study demonstrated any significant association between BMI and PIF occurence.⁸ In our study, BMI values were nearly identical in patients with and without PIF. This finding could be partly explained by global data reporting troubling increases in overweight and obesity in the last decades.²⁰ Moreover, global data report decreased age-standardized mineral bone densities in the last decades.²⁰

Some previous studies have suggested an influence of RT dose and volume of irradiated tissue or chemotherapy cycles on radiation-related bone outcomes.^21,22 We were unable to support these associations in our cohort. Thus, we did not find any statistically significant differences in the above-mentioned factors. One possible explanation is that all our patients underwent highly standardized treatment protocols using advanced RT techniques, which may have reduced variability in dose-related effects. However, in our previous retrospective study, comparing patients treated for cervical and endometrial cancer, we similarly did not observe any association between RT dose and PIF occurrence, although RT doses differed significantly between the study subgroups.⁸

Limitations

Our study has significant limitations. Although prospective in design, it is a single-center study. Our oncology center is relatively small, and there is no chance for more subjects to be included in the future. The long-term imaging follow-up should help and could also refine the results. Our patients were followed for a minimum of 1 year, and the maximum follow-up was 5 years. Also, comparing CT L1 attenuation values with DEX measurement should be very helpful and could strengthen the evidence. Moreover, for L1 attenuation measurement, we used a standard pretreatment CT scan that is excellent for morphological imaging but not ideal for material quantification. Photon-counting CT or dual energy CT are considered as powerful diagnostic tools with material decomposition capabilities that are able to determine qualitative and quantitative information.^23,24 Finally, the study lacked detailed radiation dosimetry, including site-specific dosimetry to individual skeletal structures, which could further clarify the relationship between local radiation exposure and bone injury.

Conclusion

Despite the highly standardized treatment using advanced RT techniques, the incidence of radiation-related PIF remained high, reaching nearly 27% in our patients’ cohort. Moreover, Bayesian modeling indicated that an estimated probability of PIF for a 60-year-old woman increased to 34% and to 58% for a 70-year-old woman. We concluded that, in addition to age, the lower pretreatment CT L1 attenuation values were strong predicting factors for PIF occurrence that reflect osteoporosis status. Pretreatment L1 CT attenuation values of 100 HU were associated with an estimated 51% probability of PIF. Patients with pretreatment L1 CT attenuation values below 150 HU and older than 50 years appear to be at increased risk of radiation-related PIF and may benefit from closer clinical and imaging surveillance.

Supplemental Material

sj-docx-1-whe-10.1177_17455057261426799 – Supplemental material for Pretreatment computed tomography L1 attenuation values: Easy reaching predicting factor for radiation-related bone insufficiency fractures in females treated for advanced cervical cancer (prospective study)

Supplemental material, sj-docx-1-whe-10.1177_17455057261426799 for Pretreatment computed tomography L1 attenuation values: Easy reaching predicting factor for radiation-related bone insufficiency fractures in females treated for advanced cervical cancer (prospective study) by Hana Malikova, Klaudia Graffneter, Karin Kremenova, Romana Burgetova, Viktor Laskov, Jiri Lukavsky and Lukas Rob in Women's Health

Footnotes

Acknowledgements

None.

ORCID iDs

Hana Malikova

Klaudia Graffneter

Karin Kremenova

Romana Burgetova

Viktor Laskov

Jiri Lukavsky

Lukas Rob

Ethical Considerations

The study was conducted according to guidelines of the Declaration of Helsinki and approved by the Institutional Ethics Committee of Faculty Hospital Kralovske Vinohrady, Prague, Czech Republic (EK-VP/22/00/2019; EK-VP/61/0/2020).

Consent to Participate

All the participants provided written informed consent and agreed to data publication in an anonymous form.

Author contributions

Hana Malikova: Conceptualization; Methodology; Supervision; Writing – original draft; Writing – review & editing; Resources; Data curation; Validation; Investigation; Formal analysis; Project administration.

Klaudia Graffneter: Writing – original draft; Investigation; Writing – review & editing.

Karin Kremenova: Investigation; Writing – original draft; Writing – review & editing.

Romana Burgetova: Writing – original draft; Investigation; Writing – review & editing.

Viktor Laskov: Investigation; Writing – original draft; Writing – review & editing.

Jiri Lukavsky: Formal analysis; Software; Writing – original draft; Writing – review & editing.

Lukas Rob: Writing – original draft; Writing – review & editing; Supervision.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Supplemental material

Supplemental material for this article is available online.

References

Sung

Ferlay

Siegel

, et al. Global Cancer Statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2021; 71: 209–249.

Bhatla

Aoki

Sharma

, et al. Cancer of the cervix uteri: 2021 update. Int J Gynaecol Obstet 2021; 155(Suppl 1): 28–44.

Sakyi

Miller

Coleman

, et al. Effects of radiotherapy upon bone structure-strength relationships vary with sex and fractionation of dosing. Iowa Orthop J 2023; 43: 77–86.

Haque

Hossen

. Insights into pelvic insufficiency fracture following pelvic radiotherapy for cervical cancer: a comparative review. BMC Womens Health 2024; 24: 306.

Chung

Lee

Y-K

Yoon

B-H

, et al. Pelvic insufficiency fractures in cervical cancer after radiation therapy: a meta-analysis and review. In Vivo 2021; 35: 1109–1115.

Malikova

Burghardtova

Fejfarova

, et al. Advanced cervical cancer in young women: imaging study of late and very late radiation-related side effects after successful treatment by combined radiotherapy. Quant Imaging Med Surg 2021; 11: 21–31.

Nadova

Burghardtova

Fejfarova

, et al. Late radiation-related toxicities in patients treated for early-stage cervical carcinoma by surgery and adjuvant radiotherapy: a retrospective imaging study. Pathol Oncol Res 2021; 27: 1609915.

Malikova

Nadova

Reginacova

, et al. Radiation-related fractures after radical radiotherapy for cervical and endometrial cancers: are there any differences? Diagnostics (Basel) 2024; 14: 810.

Jang

Graffy

Ziemlewicz

, et al. Opportunistic osteoporosis screening at routine abdominal and thoracic CT: normative L1 trabecular attenuation values in more than 20 000 adults. Radiology 2019; 291: 360–367.

10.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008; 61: 344–349.

11.

Bürkner

. brms: An R Package for Bayesian Multilevel Models Using Stan. J Stat Softw 2017; 80: 1–28.

12.

Bürkner

. Advanced Bayesian Multilevel Modeling with the R Package brms. R J 2018; 10: 395–411.

13.

Bürkner

. Bayesian Item Response Modeling in R with brms and Stan. J Stat Softw 2021; 100: 1–54.

14.

Pickhardt

Pooler

Lauder

, et al. Opportunistic screening for osteoporosis using abdominal computed tomography scans obtained for other indications. Ann Intern Med 2013; 158: 588–595.

15.

Salcedo

Sood

Jhingran

, et al. Pelvic fractures and changes in bone mineral density after radiotherapy for cervical, endometrial, and vaginal cancer: a prospective study of 239 women. Cancer 2020; 126: 2607–2613.

16.

Kurrumeli

Oechsner

Weidenbächer

, et al. An easy way to determine bone mineral density and predict pelvic insufficiency fractures in patients treated with radiotherapy for cervical cancer. Strahlenther Onkol 2021; 197: 487–493.

17.

Tokumaru

Toita

Oguchi

, et al. Insufficiency fractures after pelvic radiation therapy for uterine cervical cancer: an analysis of subjects in a prospective multi-institutional trial, and cooperative study of the Japan Radiation Oncology Group (JAROG) and Japanese Radiation Oncology Study Group (JROSG). Int J Radiat Oncol Biol Phys 2012; 84: e195–e200.

18.

Al-Bashaireh

Haddad

Weaver

, et al. The effect of tobacco smoking on bone mass: an overview of pathophysiologic mechanisms. J Osteoporos 2018; 2018: 1206235.

19.

Schmeler

Jhingran

Iyer

, et al. Pelvic fractures after radiotherapy for cervical cancer: implications for survivors. Cancer 2010; 116: 625–630.

20.

GBD 2021 Risk Factors Collaborators. Global burden and strength of evidence for 88 risk factors in 204 countries and 811 subnational locations, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet 2024; 403: 2162–2203.

21.

Mori

Okonogi

Matsumoto

, et al. Effects of dose and dose-averaged linear energy transfer on pelvic insufficiency fractures after carbon-ion radiotherapy for uterine carcinoma. Radiother Oncol J 2022; 177: 33–39.

22.

Misra

Lal

Kumar Ep

, et al. Comparative assessment of late toxicity in patients of carcinoma cervix treated by radiotherapy versus chemo-radiotherapy—minimum 5 years follow up. Cancer Treat Res Commun 2018; 14: 30–36.

23.

Bhattarai

Tanaka

Yeung

AWK

, et al. Photon-counting CT material decomposition in bone imaging. J Imaging 2023; 9: 209.

24.

Fervers

Weisthoff

, et al. Dual-energy CT, virtual non-calcium bone marrow imaging of the spine: an AI-assisted, volumetric evaluation of a reference cohort with 500 CT scans. Diagnostics (Basel) 2022; 12: 671.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB