Abstract
Objectives:
To evaluate the effectiveness of follow-up with non-enhanced CT (NECT) in patients with breast cancer.
Methods:
The present retrospective study included 1396 patients with breast cancer. Group A included patients with no metastasis to evaluate the diagnostic performance of NECT in detecting newly developed metastasis. Group B included patients with known hepatic metastasis to evaluate the accuracy of NECT for the assessment of hepatic metastasis.
Results:
Group A included 895 patients (mean age 52.8 years). Among them, 145 patients had 160 metastases. The per-patient sensitivities for diagnosing newly developed metastasis were 68.3 and 53.8% according to the two reviewers, while the per-lesion sensitivities were 89.4 and 85.0%. Sensitivities for bone metastasis were 98.9 and 95.9%, while sensitivities for hepatic metastasis were 73.7 and 68.4%. In group B, the accuracy of hepatic metastasis response evaluation according to the RECIST criteria was 70.8% for reviewer 1 and 63.8% for reviewer 2.
Conclusions:
NECT showed inadequate diagnostic performance in detecting newly developed metastasis and in evaluating the response of hepatic metastasis. However, NECT can be utilized as a follow-up modality in patients with decreased renal function or hypersensitivity to iodinated contrast media.
Advances in knowledge:
The risk of side effects of contrast media should be considered as important when NECT can be utilized as a follow-up modality in decreased renal function patients.
Introduction
The incidence of breast carcinoma has been increasing over the last decade, partly due to increased detection as a result of improvements in screening programs. 1 Breast cancer is the most common cancer among females worldwide. It is also a common cause of metastasis including hepatic and bone metastasis detected on abdominopelvic computed tomography (APCT) scan. 2 Detection of distant metastasis is clinically important in these patients for the initial staging as well as for the assessment of response to chemotherapeutic regimens.
Contrast-enhanced computed tomography (CECT) is widely used as the primary imaging modality for the evaluation of distant metastasis. The reported overall sensitivity of recent multi-detector computed tomography (CT) systems for detecting hepatic metastasis was approximately 72–83%. 3,4
When CECT is used as the primary follow-up imaging modality in breast cancer patients, the risk of side effects due to the contrast media (CM) should be considered as important as the risk of radiation dose while conducting CT examinations. 5 Although current iodinated contrast media have a relatively lower risk compared to their predecessors, the incidence of adverse outcomes still approaches 3%. 6 These risks include controversial risks such as contrast-induced nephropathy and non-specific risks such as febrile sensation, headache, nausea, and vomiting. However, the incidence of severe life-threatening adverse risk associated with iodinated contrast media is thought to be approximately one of every 1000 uses with mortality rates of around one of every 100 000 uses. 7 Iodinated contrast media plays an important role in many imaging modalities. Thus, contrast studies exhibit superior sensitivity and specificity compared to non-contrast studies in many cases, partly due to the improved contrast resolution. The utility of iodinated contrast has been assessed within the context of many disease processes, patient presentations, and clinical settings including both acute 8 and non-acute/oncological indications. 9
A previous study 10 reported that the diagnostic results of non-contrast-enhanced CT (NECT) and CECT were consistent with respect to the status of hepatic metastasis in 71% of the cases of colon cancer. However, the aforementioned study was performed before the establishment of multi detector CT or advanced injection protocols for contrast media. According to the recent American College of Radiology appropriateness criteria, NECT is usually not appropriate as a surveillance tool for follow up of the primary malignancy, but can be potentially indicated if patients cannot receive iodinated contrast media. 11 According to the Response Evaluation Criteria In Solid Tumors (RECIST, v.1.1), the appropriateness of NECT for the evaluation of patients who cannot receive contrast media should be carefully discussed with a radiologist with due consideration to the tumor type and the anatomic location of the lesions either before enrollment or after obtaining baseline contrast study. Considering the rapidly increasing prevalence of chronic renal disease in the aging population, the diagnostic performance of NECT for the evaluation of metastasis should be re-evaluated, since the number of patients indicated for CT examination is increasing. While most centers routinely use CECT in follow-up for breast cancer patients, many clinicians want to use or are using NECT at follow-up to evaluate the less common abdominopelvic metastasis in patients with confirmed diagnosis of cancer including breast cancer. 12,13 However, no studies have been conducted regarding the usefulness of NECT in the follow-up of abdominopelvic metastasis in breast cancer patients. Hence, the primary purpose of this study was to evaluate the performance for detecting newly developed abdominopelvic metastasis on NECT. The second purpose was to evaluate the tumor response in known hepatic metastasis using NECT in the breast cancer patient.
Methods and materials
The institutional review board approved the present retrospective study and informed consent was waived.
Patients
Initially, we included 1396 consecutive APCT scans performed in patients aged above 18 years for the follow-up evaluation of previously diagnosed breast cancer at our medical center between January 2005 and December 2014. The inclusion criteria were as follows: (1) patients with histopathological proven breast cancer; (2) patients who had undergone contrast-enhanced APCT including portal venous phase (PVP) imaging (CT1), and (3) patients who had undergone regular follow-up at our medical center for at least 12 months after CT1 with contrast-enhanced APCT including non-contrast imaging and PVP (CT2), which were performed within 12 months. The patients who had undergone locally treatment for hepatic metastasis such as radiofrequency ablation (RFA) or percutaneous ethanol injection (PEI) between the two CT scans were excluded.
Finally, 1073 patients were included in this study, more detailed demographical and histological information are provided in Supplementary Table 1.
Patients were divided into two groups: group A and group B. Group A included asymptomatic patients without metastasis in CT1 regardless of whether or not there was metastasis in CT2 for analyze diagnostic performance of NECT for detection of newly developed metastasis. Group B included patients with known hepatic metastasis on CT1 for evaluate the diagnostic performance of NECT according to the RECIST v.1.1 criteria. 11 Among patients in group B, those who did not have definite target lesions on CT1 or CT2 (all hepatic metastatic lesions <10 mm in size or non-visualized disseminated bone metastases) were also excluded.
Imaging technique
CT examinations were performed using one of the five different scanners available at our institution, which included SOMATOM Sensation 64, SOMATOM Definition, SOMATOM Force, SOMATOM AS-Edge, and SOMATOM Definition Flash (Siemens Healthcare, Forchheim, Germany) available at our medical center. CT examinations were performed using a spiral technique with 3–5 mm collimation. A non-contrast image was obtained before the administration of contrast media.The scans were acquired 60–75 s after initiating intravenous injection of 100 ml of non-ionic contrast medium (iohexol, Omnipaque 300; Amersham Health, Princeton, NJ) at a rate of 2–3 mL second−1 using a power injector via the antecubital vein. Axial sections of 5 mm thickness were reconstructed, sent for reporting, and archived.
Performance of non-contrast CT scan
Two experienced abdominal radiologists (experience of 15 years and 5 years, respectively, in abdominal imaging) independently reviewed the NECT images of session 1 (group A) and session 2 (group B) in separately. The reviewers were aware of presence or absence of metastasis and the site of metastasis in both the groups. However, they did not know any other clinical information. In the session 1, the reviewers analyzed the CT2 whether there was developed metastasis compared with CT1. For the developed metastasis, the reviewers were marked the locations of lesion and measured the largest diameter going up to five lesions according to size. Subsequently, the reviewers were graded the confidence level of the lesion on a four-point scale (1: no lesion, 2: probably no lesion, 3: probably lesion, 4: definitely lesion). Lesions with confidence levels of 3 or 4 were regarded as metastases. Significant lymph node involvement was defined as lymph nodes > 1 cm in size. In group B, the reviewers measured the two largest metastases (target lesions) in CT1 and CT2 images and to determine the response after chemotherapy based on the RECIST v.1.1 criteria. 11 The response was categorized as complete response: disappearance of all target lesions, partial response: at least a 30% decrease in the sum of the diameters of the target lesions, progressive disease: at least a 20% increase in the sum of the diameters of the target lesions or the appearance of one or more new lesions, or stable disease: neither sufficient shrinkage to qualify as partial response nor sufficient increase to qualify as progressive disease.
Reference standard
The study coordinators had a clinical experience of 15 years and 10 years, respectively in abdominal radiology. Decisions regarding the presence or the absence of metastasis were made in consensus based on CECT; ultrasound (US); magnetic resonance imaging (MRI); positron emission tomography (PET)-CT scan; follow-up US, CT, and MRI; and pathological reports of the excision or biopsy specimens. The confirmation of malignancy was based on pathology or image surveillance. In group A, 160 metastatic sites were identified in 145 patients based on the following criteria: 1) surgery (2 sites in two patients with skin and adrenal gland metastasis), 2) needle biopsy (41 sites in 36 patients), and 3) tumor growth observed on cross-sectional follow-up imaging (117 sites in 107 patients). All patients who underwent surgery or biopsy were followed-up with CECT scan for at least 6 months. Follow-up was extended in case of development of additional recurrent lesions. In the absence of histopathological data, metastasis was confirmed when the lesion show the typical image findings of metastasis 14,15 on at least two image modalities and when observed the interval growth in the longest axial diameter was at least 20% on follow-up imaging. The mean follow-up interval was 11.2 ± 3.2 months (range: 5–18 months). The increased size lesions in the follow-up image were excluded from the benign lesion group, even although there was a high possibility of benign lesions such as hemangioma. In 750 patients who had no metastasis, the absence of metastasis was confirmed on follow-up image studies performed after at least 6 months after CT1.
Statistical analysis
Diagnostic performances were analyzed using a frequency table and the generalized estimating equation approach. Interobserver agreement was evaluated using Cohen’s κ statistic. The degree of agreement was categorized according to the κ values as almost perfect (between 0.81 and 1.00), substantial (between 0.61 and 0.80), moderate (between 0.41 and 0.60), fair (between 0.21 and 0.40), or poor (between 0.00 and 0.20). Data were analyzed using SPSS Statistics v.18.0 (SPSS Inc., Chicago, IL, USA) Statistical significance was set at p < 0.05 for all tests.
Results
Patients
After a review of patient’s medical records, 1396 patients were identified from January 2005 to December 2014. Among these, 212 patients were excluded after the review of CT images due to the absence of non-contrast images or PVP images on follow-up APCT; 106 patients were excluded, as no follow-up APCT was available within the period of study; and five patients were excluded, as RFA, PEI or surgery was performed between the two CT scans. Among the remaining 1073 patients, 895 patients (mean age: 52.8 years, range: 21–86 years) were classified into group A. In group A, 145 patients had developed metastases on CT2. Newly developed metastases included 160 lesions (98 bone metastases, 57 hepatic metastases, three lymph node metastases, one adrenal gland metastasis, and one skin metastasis). Among these, five lesions including lymph node, adrenal gland, and skin metastases were categorized into the “others” group. Thus, the analysis included three groups, namely bone, liver, and others.
Since additional 18 patients were excluded due to the absence of target lesions on CT1, 160 out of 178 patients were finally included in group B (mean age: 58.2 years, range: 25–89 years) (Figure 1). Group B consisted of of 102 bone metastases, 55 hepatic metastases, two lymph node metastases, and one adrenal gland metastasis.
Figure 1.
Flow chart of patients’ inclusion and subgrouping.
Side effects of contrast medium
Twenty-eight patients out of 1055 (2.7%, including both groups) experienced side-effects of CM. Mild hypersensitivity reactions (n = 25) consisted of flushing or urticarial, skin rashes, rhinorrhea, pruritus, coughing, nausea, dizziness, and diaphoresis. Moderate to severe (n = 3) reactions included headache, diffuse urticaria, persistent vomiting, mild bronchospasm, abdominal cramps, palpitations, tachycardia, laryngeal edema, and facial edema. There were no deaths related to the injection of CM. Infusion of CM was ceased and treatment with antihistamines was initiated immediately on detection of a reaction. Patients with moderate to severe side-effects were treated immediately with intravenous fluids, oxygen, adrenaline, and in addition to antihistamines with/ without steroid.
Twenty-three patients exhibited improvement in symptoms within six hours of the treatment. Thus, no additional treatment or hospitalization was needed for these patients. However, in the remaining five patients, hospitalization was required for observation. In case of inpatients, the length of hospitalization was extended. Among these, no patients were admitted to the intensive care unit and all were discharged within 3 days without further examination or treatment.
In our study, the rate of contrast extravasation during CT was 0.28% (3/1055). And there were no contrast induced nephropathy in the patients included in this study.
Imaging analysis
Group A comprised of 160 metastases in 145 patients. In the per-patient analysis of group A, sensitivity and specificity were 68.3 and 96.0%, respectively, according to reviewer 1 and 53.8% and 95.2, respectively according to reviewer 2 (Table 1). A statistically significant difference was observed in sensitivity (p = 0.038), but not in specificity (p = 0.081). Interobserver agreement between the reviewers was substantial [κ = 0.68, 95% confidence interval (CI): 0.51–0.75]. In the per-lesion analysis, sensitivity was 89.4% according to reviewer 1 and 85.0% according to reviewer 2 (Table 2, Figure 2). Interobserver agreement was substantial too [κ = 0.72, 95% CI (0.59–0.79)]. Sensitivity for bone metastasis was 98.9% according to reviewer 1 and 95.9% according to reviewer 2. Sensitivity for hepatic metastasis was 73.7% according to reviewer 1 and 68.4% according to reviewer 2. Sensitivity for other metastases was 80.0% according to reviewer 1 and 60.0% according to reviewer 2. Sensitivities in the per-lesion analysis and in the subgroup analysis were significantly higher according to reviewer one than according to reviewer 2 (p = 0.041). Sensitivity for hepatic metastasis was lower than that for bone metastasis according to both reviewer 1 and reviewer 2. Among the 57 developed hepatic metastases, 41 nodules were larger than 15 mm in size. Diagnostic performance of lesions larger than 15 mm was significantly higher than that for all hepatic metastases according to both reviewers (92.7 and 73.7% according to reviewer 1, 85.4% and 68.4% according to reviewer 2) (Supplementary Table 2). In group B, the accuracy of hepatic metastasis response evaluation was 70.8% according to reviewer 1 and 63.8% according to reviewer 2 (Table 3, Figure 3). A significant difference was observed in diagnostic accuracy between the two reviewers. Moreover, there was a tendency toward over diagnosis in case of reviewer 2 (Table 3). Interobserver agreement was substantial [κ = 0.75, 95% CI (0.58–0.82)].
Table 1.
Per-patient analysis of diagnostic performance in the detection of newly developed metastasis (group A)
| Per-patient | SEN (95% CI) | SPEC (95% CI) | ACC (95% CI) | PPV (95% CI) | NPV (95% CI) |
|---|---|---|---|---|---|
| Reviewer 1 | 68.3 (60.0–75.8) | 96.0 (94.3–97.3) | 91.5 (89.5–93.2) | 76.7 (69.6–82.7) | 93.9 (92.5–95.2) |
| Reviewer 2 | 53.8 (45.3–62.1) | 95.2 (93.4–96.6) | 88.5 (86.2–90.5) | 68.4 (60.4–75.5) | 91.4 (89.9–92.3) |
| p value | 0.038 a | 0.081 | 0.065 | 0.059 | 0.120 |
ACC, Accuracy; CI, Confidence interval; NPV, Negative predictive value; PPV, Positive predictive value; SEN, Sensitivity; SPEC, Specificity.
Statistically significant.
Table 2.
Per-lesion analysis of diagnostic performance in the detection of newly developed hepatic metastasis (group A)
| SEN (95% CI) | SPEC (95% CI) | ACC (95% CI) | PPV (95% CI) | NPV (95% CI) | ||
|---|---|---|---|---|---|---|
| All lesions | Reviewer 1 | 89.4 (83.5–93.7) | 94.5 (93.6–95.4) | 94.2 (93.3–95.1) | 50.89 (46.6–55.1) | 99.3 (98.9–99.6) |
| Reviewer 2 | 85.0 (78.5–90.2) | 94.8 (93.8–95.6) | 94.2 (93.2–95.0) | 50.8 (46.3–55.2) | 99.0 (98.6–99.3) | |
| p value | 0.041 a | 0.092 | 0.103 | 0.115 | 0.072 | |
| Bone | Reviewer 1 | 98.9 (94.4–99.9) | 87.7 (85.2–89.9) | 88.9 (86.7–90.9) | 49.7 (45.1–54.4) | 99.9 (99.0–99.9) |
| Reviewer 2 | 95.9 (89.9–98.9) | 88.1 (85.6–90.3) | 88.9 (86.7–90.9) | 49.8 (44.9–54.6) | 99.4 (98.5–99.8) | |
| p value | 0.036 a | 0.124 | 0.185 | 0.094 | 0.107 | |
| Liver | Reviewer 1 | 73.7 (60.3–84.5) | 95.8 (94.2–97.1) | 94.4 (92.7–95.8) | 54.6 (45.6–63.2) | 98.2 (97.2–98.8) |
| Reviewer 2 | 68.4 (54.8–80.0) | 96.7 (95.2–97.8) | 94.9 (93.2–96.2) | 58.2 (48.2–67.6) | 97.8 (96.8–98.5) | |
| p value | 0.044 a | 0.087 | 0.107 | 0.068 | 0.085 | |
| Others | Reviewer 1 | 80.0 (28.4–99.5) | 99.4 (98.7–99.8) | 99.3 (98.6–99.8) | 44.4 (23.1–68.0) | 99.8 (99.3–99.9) |
| Reviewer 2 | 60.0 (14.7–94.8) | 98.9 (98.1–99.5) | 98.8 (97.8–99.4) | 25.0 (11.3–46.7) | 99.8 (99.3–99.9) | |
| p value | 0.038 a | 0.174 | 0.215 | 0.024 a | 0.119 |
ACC, Accuracy; CI, Confidence interval; NPV, Negative predictive value; PPV, Positive predictive value; SEN, Sensitivity; SPEC, Specificity.
Statistically significant.
Figure 2.
Representative case of a 53-year-old female who underwent right mastectomy due to breast cancer, histopathologically confirmed as invasive ductal carcinoma (grade II, ER negative, PR negative). On follow-up CT, newly developed metastasis was observed in the right hemiliver (arrows). (a, b) Neither of the reviewers could detect these lesions on NECT.
Table 3.
Accuracy of hepatic metastasis response evaluation according to the RECIST 1.1 criteria
| Accurate diagnosis (95% CI) | Underdiagnosis (95% CI) | Overdiagnosis (95% CI) | |
|---|---|---|---|
| Reviewer 1 | 70.8 (63.1–80.5) | 17.5 (12.1–21.7) | 11.7 (8.5–17.2) |
| Reviewer 2 | 63.8 (58.1–71.8) | 19 (12.3–25.8) | 17.2 (12.1–22.8) |
| p value | 0.040 a | 0.122 | 0.034 a |
CI, Confidence interval; RECIST, Response evaluation criteria In solid tumors.
Statistically significant.
Figure 3.
Representative case of a 72-year-old female who had known liver metastasis associated with breast cancer, histopathologically confirmed as invasive ductal carcinoma (grade II, ER positive, PR negative). (a) On portal venous Phase imaging, hepatic metastasis was observed in segment 8 of the liver. (b) The size of the lesion appeared similar in the non-enhanced CT image. The lesion was 31 mm in size in contrast enhanced CT (a) as well as in non-enhanced CT (b) images. (c, d) On follow-up CT after 3 months, the lesion had decreased in size. It was 19 mm in size (57% interval decrease) on contrast-enhanced CT, which suggested partial response of disease (c). However, on non-enhanced CT, the mass was 25 mm in size (19% interval decrease), which suggested stable disease (d).
Clinical impact of non-contrast-enhanced CT scan
Metastasis was missed on NECT in 46 patients form group A. Among these, only eight had missed metastasis without metastasis in other body parts such as lung and bone. Therefore, the stage of the breast cancer was underdiagnosed. These patients had liver (n = 5), bone (n = 2), and lymph node (n = 1) metastases. Among these, chemotherapy regimen was changed in four patients and one patient underwent additional radiation therapy for sacral bone metastasis. The remaining three patients were additionally treated with the remaining cycles of chemotherapy.
Among the 160 patients in group B, 28 were underdiagnosed and 19 were overdiagnosed on NECT. Particularly, 16 out of 28 underdiagnosed patients were scheduled to undergo chemotherapy when diagnosed with NECT alone.
Discussion
Our results revealed that the diagnostic sensitivities of NECT for the detection of newly developed breast cancer metastasis were 68.3% (reviewer 1) and 53.8% (reviewer 2) in the per-patient analysis and 89.4% (reviewer 1) and 85.0% (reviewer 2) in the per-lesion analysis. The difference in sensitivity between the two reviewers was statistically significant. This difference might be attributed to the difference in experience between the two reviewers. 16,17 Among the bone, liver, and other metastasis groups, sensitivity for bone metastasis was higher than 95% according to both reviewers. However, sensitivity for liver metastasis was lower than 75% (73.7 and 68.4% according to reviewer 1 and 2, respectively).
Similar to our results, previous studies have identified bone as the most common site of distant metastasis in breast cancer among the organs included in APCT. 18 In contrast, liver was observed to be the least common site of metastasis. According to a study conducted by Abigail et al 19 on 1,754 patients with early-stage breast cancer, bone metastasis accounted for more than 40%, while liver metastasis accounted for only 7% of the cases of metastasis after breast conservation treatment.
APCT exhibits higher sensitivity for the detection of bone metastasis than for the detection of soft tissue metastasis including liver metastasis. This is probably due to the difference in tissue contrast on CT scan. CT provides excellent resolution of trabecular bone and cortical bone and is the first choice of imaging modality. According to adjust the window width and level to acquired images, and to view the skeleton using multiplanar reformatted images serves to maximize the conspicuity of bone lesions and results in a higher sensitivity in detecting osteosclerotic as well as osteolytic metastases. 20 Therefore, CT can demonstrate bone marrow metastases before bone destruction occurs, which helps in earlier diagnosis, improved prognosis, and prevention of complications. 21 Sensitivity for bone metastasis observed in our study was higher than that reported by Piccardo et al. (77%). 22 This difference might be due to the fact that our study procedure detected developed lesions after comparison with the previous APCT.
The accuracy of NECT for the detection of hepatic metastasis is not satisfactory compared to its accuracy for the detection of bone metastasis. Due to the inherently low contrast resolution of the soft tissue including liver, optimal contrast enhancement is indispensable for accurate diagnosis of focal hepatic lesion. According to previous studies, at least 50 HU of hepatic enhancement is required to obtain high diagnostic quality CT images of liver. 23 Bernard et al identified no significant difference in the diagnosis of focal hepatic lesions between NECT and CECT, 24 but a recent study has reported that PVP showed better diagnostic performance than NECT. PVP could detect 85% of the liver metastases, but only 61% were detected on NECT. Moreover, there were no patients in whom all hepatic metastatic lesions were missed on PVP. 25 Our research showed similar results. Approximately 70% of the metastatic lesions were diagnosed on NECT by both reviewers in our study. Based on these results, follow-up with NECT in patients who might develop new hepatic metastatic lesions might be inadequate.
The per-lesion diagnostic sensitivity for hepatic metastasis increased significantly for both reviewers in the diagnosis of metastatic lesions larger than 1.5 cm (from 73.7 to 92.7% in reviewer 1 and from 68.4 to 85.4% in reviewer 2). Thus, NECT was highly accurate in detecting clinically meaningful liver metastasis, avoiding the need for additional medical resources and economic burden that is usually associated with accurate diagnosis of liver lesions that are too small to be characterized. 26,27
NECT has limitations in the detection of small hepatic metastases, but shows high accuracy in bone metastasis, which is the most common metastasis in breast cancer. Hence, it can be an alternative tool for the evaluation of metastasis in patients with impairment or renal function or in those with low probability of hepatic metastasis.
US is widely used as follow-up imaging modality, as it is easily available and harmless to the patients. However, sensitivity of conventional US for the detection of liver metastasis was only 40–55%. 3,28 Moreover, bone metastasis is also difficult to detect using US. Therefore, it is not suitable as the only follow up imaging modality for the patients with potential for developing metastasis. CEUS is possible alternative tool that show diagnostic performance similar to that of CECT. 28 However, accessibility is limited, as CEUS requires dedicated software. Contrast-enhanced MRI had excellent diagnostic performance for hepatic metastasis with a sensitivity of approximately 76–88%. 3,29 However, patients with impaired renal function are considered to have a relative contraindication for contrast-enhanced MRI(CE MRI) due to the possibility of nephrogenic systemic fibrosis. According to previous meta-analysis, non-contrast MRI show comparable sensitivity with CE MRI. 29 Thus, MRI should be performed for each clinically suspected area in most of the medical centers. However, hospitals that can implement whole-body MRI are limited. Furthermore, the current long examination time, high cost, and variable global availability of MRI interfere its wide use as the first-line imaging modality for the assessment of hepatic metastasis. 25,30 18F-fluorodeoxyglucose (FDG) PET-CT can be another option for the evaluation of hepatic metastasis without iodinated CM. A recent study demonstrated that PET-CT shows good diagnostic performance for the detection of metastasis and can prevent unnecessary surgery in cases of bone metastasis as well as hepatic metastasis. 31 However, the sensitivity of PET-CT has been observed to decrease markedly after chemotherapy. Most of breast cancer patients with hepatic metastasis undergo chemotherapy before and/or after the surgery. Thus, PET-CT is unlikely to completely replace CECT or MRI. NECT has limitations in the detection of small hepatic metastases, but shows high accuracy in bone metastasis, which is the most common metastasis in breast cancer. Hence, it can be an alternative tool for the evaluation of metastasis in patients with impairment or renal function or in those with low probability of hepatic metastasis.
In terms of diagnostic performance for analyzing treatment response of hepatic metastasis according to the RECIST 1.1 criteria, only 63.8–70.8% of the patients were assessed correctly by NECT. This finding is consistent with the finding of a previous study that revealed that approximately 71% of the patients showed the similar results between NECT and CECT in the evaluation of hepatic metastasis associated with gastrointestinal cancer. 10,12 Jee et al 12 demonstrated that the sizes of hepatic metastasis of gastrointestinal cancer measured on NECT were smaller than the sizes measured on CECT for all lesions, significantly. The misdiagnosis or the discrepancy in size measurement may be attributed to the fact that low-contrast resolution in NECT is inferior to that in CECT. In NECT, the central necrotic portion of the tumor could be differentiated from the surrounding normal hepatic parenchyma, but the peripheral viable portion could not be differentiated, resulting in misdiagnosis of hepatic metastasis. Our study results suggested that overestimation or underestimation of treatment response may occur in approximately one-third of the patients. According to The RECIST 1.1 criteria, an optimal and consistent CM injection protocol is important for accurate assessment of patients. If the iodinated CM administration is contraindicated, other alternative imaging modality such as NECT, US, or MRI should be selected according to the anatomical location of the tumor or tumor type in the each patient. 11 We evaluated tumor response only in cases of hepatic metastases and not in bone metastases, as CT measurements of soft tissue diseases are incorporated into the RECIST 1.1 criteria. However, CT shows poor performance when used to detect and assess the response to therapy for bone metastases. 32 Bone metastases are considered non-evaluable according to the RECIST 1.1 criteria unless there is an associated soft tissue component, negating the trial entry of many patients. 11
The present study has several limitations. This study was a retrospective study. Therefore, the patient characteristics were heterogeneous. Patients had different breast cancer stages, proceed different post-operative treatments (Hormone therapy, chemotherapy, radiation therapy), and CT was performed at different intervals.
However, our objective was to evaluate the diagnostic performance of NECT in the detection and evaluation of metastasis in breast cancer. Hence, heterogeneous patient characteristics reflected actual clinical situations and might not have influenced the results significantly. In the present study, most of the metastatic lesions were not confirmed histologically. However, two experienced radiologists analyzed all images, laboratory results and clinical information, and imaging obtained over a follow-up period of at least 1 year to confirm the reference standards. We believe that this method is amply accurate for the diagnosis and evaluation of metastasis. And last, this study was a comparison of pre-contrast scan images and post contrast scan images included in one CT scan. This increased the radiation dose. However, this study was conducted not as “prospective” but as “retrospective”. Breast surgeon or oncologist at this medical center use pre and post contrast CT scan for metastasis evaluation in breast cancer patient. Currently, radiologists, including the authors, are recommending follow-up to post-contrast CT scan only to clinicians and the prescriptions are being revised.
In conclusion, NECT showed inadequate diagnostic performance in the detection of newly developed metastasis (especially hepatic metastasis) as well as response evaluation of hepatic metastasis after treatment according to the RECIST 1.1 criteria in breast cancer patients. Therefore, CECT with proper preparations 33 or alternative imaging modalities such as MRI and CEUS should be considered in highly suspected hepatic metastasis. However, if clinicians are aware of the limitations of NECT, it can be utilized as a follow-up modality in patients with impaired renal function or hypersensitivity to iodinated CM. Moreover, NECT is useful for the diagnosis of bone metastasis, as its accuracy is similar to that of CECT for bone metastasis, which is the most common metastatic site in breast cancer patients.
Footnotes
Patient consent: Requirements for informed consent were waived due to the retrospective nature of the study.
Contributors: All authors contributed to and approved the final version of this manuscript. Sang Yu Nam analyzed and interpreted the data, drafted the manuscript, and performed statistical analysis. Su Joa Ahn orchestrated the study concept and design, acquired data, performed critical revision of the manuscript, and supervised the overall study. Young Rock Jang, Yong Soon Chun, Heung Kyu Park, Seung Joon Choi, Hye Young Choi, and Jeong Ho Kim acquired clinical data and assisted in the critical revision of the final manuscript.
Contributor Information
Sang Yu Nam, Email: sy.nam@gilhospital.com.
Su Joa Ahn, Email: joa0827@gmail.com.
Young Rock Jang, Email: docrock112@gmail.com.
Yong Soon Chun, Email: chunysmd@gilhospital.com.
Heung Kyu Park, Email: hgjh@gilhospital.com.
Seung Joon Choi, Email: jchoi@gilhospital.com.
Hye Young Choi, Email: choihy@gilhospital.com.
Jeong Ho Kim, Email: ho7ok7@gilhospital.com.
REFERENCES
- 1. Golubnitschaja O, Debald M, Yeghiazaryan K, Kuhn W, Pešta M, Costigliola V, et al. Breast cancer epidemic in the early twenty-first century: evaluation of risk factors, cumulative questionnaires and recommendations for preventive measures. Tumour Biol 2016; 37: 12941–57. doi: 10.1007/s13277-016-5168-x [DOI] [PubMed] [Google Scholar]
- 2. Alberg AJ, Singh S, May JW, Helzlsouer KJ, Epidemiology HKJ. Epidemiology, prevention, and early detection of breast cancer. Curr Opin Oncol 2000; 12: 515–20. doi: 10.1097/00001622-200011000-00001 [DOI] [PubMed] [Google Scholar]
- 3. Kinkel K, Lu Y, Both M, Warren RS, Thoeni RF. Detection of hepatic metastases from cancers of the gastrointestinal tract by using noninvasive imaging methods (US, CT, MR imaging, PET): a meta-analysis. Radiology 2002; 224: 748–56. doi: 10.1148/radiol.2243011362 [DOI] [PubMed] [Google Scholar]
- 4. Mainenti PP, Mancini M, Mainolfi C, Camera L, Maurea S, Manchia A, et al. Detection of colo-rectal liver metastases: prospective comparison of contrast enhanced us, multidetector CT, PET/CT, and 1.5 Tesla Mr with extracellular and reticulo-endothelial cell specific contrast agents. Abdom Imaging 2010; 35: 511–21. doi: 10.1007/s00261-009-9555-2 [DOI] [PubMed] [Google Scholar]
- 5. Beckett KR, Moriarity AK, Langer JM. Safe use of contrast media: what the radiologist needs to know. Radiographics 2015; 35: 1738–50. doi: 10.1148/rg.2015150033 [DOI] [PubMed] [Google Scholar]
- 6. Urrutia M, Macharaviaya A, Rodríguez R. Adverse reactions to contrast media for intravenous use. A comparison between ionic and nonionic media. Rev Med Panama 1995; 20(1-2): 20–4. [PubMed] [Google Scholar]
- 7. Idée J-M, Pinès E, Prigent P, Corot C. Allergy-like reactions to iodinated contrast agents. A critical analysis. Fundam Clin Pharmacol 2005; 19: 263–81. doi: 10.1111/j.1472-8206.2005.00326.x [DOI] [PubMed] [Google Scholar]
- 8. Larson DB, Johnson LW, Schnell BM, Salisbury SR, Forman HP. National trends in CT use in the emergency department: 1995-2007. Radiology 2011; 258: 164–73. doi: 10.1148/radiol.10100640 [DOI] [PubMed] [Google Scholar]
- 9. Sheafor DH, Frederick MG, Paulson EK, Keogan MT, DeLong DM, Nelson RC. Comparison of unenhanced, hepatic arterial-dominant, and portal venous-dominant phase helical CT for the detection of liver metastases in women with breast carcinoma. AJR Am J Roentgenol 1999; 172: 961–8. doi: 10.2214/ajr.172.4.10587129 [DOI] [PubMed] [Google Scholar]
- 10. Park JH, Nazarian LN, Halpern EJ, Feld RI, Lev-Toaff AS, Parker L, et al. Comparison of unenhanced and contrast-enhanced spiral CT for assessing interval change in patients with colorectal liver metastases. Acad Radiol 2001; 8: 698–704. doi: 10.1016/S1076-6332(03)80576-7 [DOI] [PubMed] [Google Scholar]
- 11. Nishino M, Jagannathan JP, Ramaiya NH, Van den Abbeele AD. Revised RECIST guideline version 1.1: what oncologists want to know and what radiologists need to know. AJR Am J Roentgenol 2010; 195: 281–9. doi: 10.2214/AJR.09.4110 [DOI] [PubMed] [Google Scholar]
- 12. Jee HB, Park MJ, Lee HS, Park M-S, Kim M-J, Chung YE. Is non-contrast CT adequate for the evaluation of hepatic metastasis in patients who cannot receive iodinated contrast media? PLoS One 2015; 10: e0134133. doi: 10.1371/journal.pone.0134133 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Semaan H, Bazerbashi MF, Siesel G, Aldinger P, Obri T. Diagnostic accuracy of non-contrast abdominal CT scans performed as follow-up for patients with an established cancer diagnosis: a retrospective study. Acta Oncol 2018; 57: 426–30. doi: 10.1080/0284186X.2017.1360512 [DOI] [PubMed] [Google Scholar]
- 14. Sica GT, Ji H, Ros PR. Ct and MR imaging of hepatic metastases. AJR Am J Roentgenol 2000; 174: 691–8. doi: 10.2214/ajr.174.3.1740691 [DOI] [PubMed] [Google Scholar]
- 15. Heindel W, Gübitz R, Vieth V, Weckesser M, Schober O, Schäfers M. The diagnostic imaging of bone metastases. Dtsch Arztebl Int 2014; 111: 741–7. doi: 10.3238/arztebl.2014.0741 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Blachar A, Barnes S, Adam SZ, Levy G, Weinstein I, Precel R, et al. Radiologists' performance in the diagnosis of acute intestinal ischemia, using MDCT and specific CT findings, using a variety of CT protocols. Emerg Radiol 2011; 18: 385–94. doi: 10.1007/s10140-011-0965-4 [DOI] [PubMed] [Google Scholar]
- 17. Lauritzen PM, Andersen JG, Stokke MV, Tennstrand AL, Aamodt R, Heggelund T, et al. Radiologist-initiated double reading of abdominal CT: retrospective analysis of the clinical importance of changes to radiology reports. BMJ Qual Saf 2016; 25: 595–603. doi: 10.1136/bmjqs-2015-004536 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Wang Z, Chen M, Pan J, Wang X, Chen X-S, Shen K-W. Pattern of distant metastases in inflammatory breast cancer - A large-cohort retrospective study. J Cancer 2020; 11: 292–300. doi: 10.7150/jca.34572 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Berman AT, Thukral AD, Hwang W-T, Solin LJ, Vapiwala N. Incidence and patterns of distant metastases for patients with early-stage breast cancer after breast conservation treatment. Clin Breast Cancer 2013; 13: 88–94. doi: 10.1016/j.clbc.2012.11.001 [DOI] [PubMed] [Google Scholar]
- 20. O'Sullivan GJ, Carty FL, Cronin CG. Imaging of bone metastasis: an update. World J Radiol 2015; 7: 202–11. doi: 10.4329/wjr.v7.i8.202 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Bäuerle T, Semmler W. Imaging response to systemic therapy for bone metastases. Eur Radiol 2009; 19: 2495–507. doi: 10.1007/s00330-009-1443-1 [DOI] [PubMed] [Google Scholar]
- 22. Piccardo A, Altrinetti V, Bacigalupo L, Puntoni M, Biscaldi E, Gozza A, et al. Detection of metastatic bone lesions in breast cancer patients: fused (18)F-Fluoride-PET/MDCT has higher accuracy than MDCT. Preliminary experience. Eur J Radiol 2012; 81: 2632–8. doi: 10.1016/j.ejrad.2011.12.020 [DOI] [PubMed] [Google Scholar]
- 23. Chung YE, Kim KW, Kim JH, Lim JS, Oh YT, Chung J-J, et al. Optimal delay time for the hepatic parenchymal enhancement at the multidetector CT examination. J Comput Assist Tomogr 2006; 30: 182–8. doi: 10.1097/00004728-200603000-00003 [DOI] [PubMed] [Google Scholar]
- 24. Berland LL, Lawson TL, Foley WD, Melrose BL, Chintapalli KN, Taylor AJ. Comparison of pre- and postcontrast CT in hepatic masses. AJR Am J Roentgenol 1982; 138: 853–8. doi: 10.2214/ajr.138.5.853 [DOI] [PubMed] [Google Scholar]
- 25. Frederick MG, Paulson EK, Nelson RC. Helical CT for detecting focal liver lesions in patients with breast carcinoma: comparison of noncontrast phase, hepatic arterial phase, and portal venous phase. J Comput Assist Tomogr 1997; 21: 229–35. doi: 10.1097/00004728-199703000-00012 [DOI] [PubMed] [Google Scholar]
- 26. Jones EC, Chezmar JL, Nelson RC, Bernardino ME. The frequency and significance of small (less than or equal to 15 Mm) hepatic lesions detected by CT. AJR Am J Roentgenol 1992; 158: 535–9. doi: 10.2214/ajr.158.3.1738990 [DOI] [PubMed] [Google Scholar]
- 27. Khalil HI, Patterson SA, Panicek DM. Hepatic lesions deemed too small to characterize at CT: prevalence and importance in women with breast cancer. Radiology 2005; 235: 872–8. doi: 10.1148/radiol.2353041099 [DOI] [PubMed] [Google Scholar]
- 28. Quaia E, D'Onofrio M, Palumbo A, Rossi S, Bruni S, Cova M. Comparison of contrast-enhanced ultrasonography versus baseline ultrasound and contrast-enhanced computed tomography in metastatic disease of the liver: diagnostic performance and confidence. Eur Radiol 2006; 16: 1599–609. doi: 10.1007/s00330-006-0192-7 [DOI] [PubMed] [Google Scholar]
- 29. Niekel MC, Bipat S, Stoker J. Diagnostic imaging of colorectal liver metastases with CT, MR imaging, FDG PET, and/or FDG PET/CT: a meta-analysis of prospective studies including patients who have not previously undergone treatment. Radiology 2010; 257: 674–84. doi: 10.1148/radiol.10100729 [DOI] [PubMed] [Google Scholar]
- 30. Kaur H, Hindman NM, Al-Refaie WB, Arif-Tiwari H, Cash BD, Chernyak V, et al. ACR Appropriateness Criteria® Suspected Liver Metastases. J Am Coll Radiol 2017; 14: S314–25. doi: 10.1016/j.jacr.2017.01.037 [DOI] [PubMed] [Google Scholar]
- 31. Moulton C-A, Gu C-S, Law CH, Tandan VR, Hart R, Quan D, et al. Effect of PET before liver resection on surgical management for colorectal adenocarcinoma metastases: a randomized clinical trial. JAMA 2014; 311: 1863–9. doi: 10.1001/jama.2014.3740 [DOI] [PubMed] [Google Scholar]
- 32. Suzuki C, Jacobsson H, Hatschek T, Torkzad MR, Bodén K, Eriksson-Alm Y, et al. Radiologic measurements of tumor response to treatment: practical approaches and limitations. Radiographics 2008; 28: 329–44. doi: 10.1148/rg.282075068 [DOI] [PubMed] [Google Scholar]
- 33. Kodzwa R. ACR manual on contrast media: 2018 updates. Radiol Technol 2019; 91: 97–100. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.