**2. Multimodality diagnostic imaging**

Diagnostic tests to detect epithelial ovarian cancer include using serum tumour marker levels correlated with imaging findings. Diagnostic imaging modalities that are frequently used to detect, stage and monitor treatment of ovarian cancers include ultrasound, computed tomography, magnetic resonance imaging and positron emission computed tomography.

#### **2.1. Serum tumour marker CA125**

Before illustrating diagnostic imaging methods for detecting OC, it is best to understand the embryology of the disease. Given that there is a remarkable morphologic and molecular heterogeneity in OC, therefore it has been postulated that ovarian cancers can be divided into Type I (indolent) and Type II (aggressive) tumours [4]. Type I is composed of low-grade serous, low-grade endometrioid, clear cell, mucinous and transitional carcinomas. These tumours are confined to the ovary at presentation and are relatively genetically stable and the majority display KRAS, BRAF and ERBB2 mutations [5]. Type II tumours include high-grade serous carcinoma (HGSC), undifferentiated carcinoma, and malignant mixed mesodermal tumours (carcinosarcoma), are highly aggressive, evolve rapidly and almost always present at an advanced stage. They display TP53 mutations in over 80% of cases and rarely harbour the mutations found in Type I tumours. Type I is suggested to be of Müllerian-type of tissues in origin, whereas Type II tumours are mesothelium in source and are suspected to originate

**Figure 1.** Diagnostic imaging algorithm for management of ovarian cancer (adapted from Suppiah et al. [1]).

from the Fallopian tubes [5].

176 Ovarian Cancer - From Pathogenesis to Treatment

CA125 is a protein that is found in greater concentrations in ovarian cancer tumour cells than in other cells of the human body. Therefore, a simple blood test, using a sample taken from a peripheral vein, makes it possible for it to be used as a marker to detect the presence of ovarian cancer. Nevertheless, it is non-specific for ovarian cancer, as raised levels may also be found in other malignancies, e.g., breast, lung, colon and pancreatic cancer as well as in benign conditions such as in endometriosis, pelvic inflammatory disease and ovarian cysts [10]. CA125 has a high positive predictive value (PPV) of >95%, but a low negative predictive value (NPV) ranging from 50 to 60%, for the detection of OC [11]. Some studies have advocated the use of a single CA125 level measurement, frequently quoting the value 35 IU/ml as a cut-off point to indicate the presence of malignancy [12, 13]. Levels above this have a good positive predictive value, however many actual cancers may have lower levels of CA125 and can be missed [14].

The National Institute for Health and Care Excellence (NICE) clinical guidelines recommend further tests to be done if a CA125 level of >35 IU/ml is detected in a woman suspected to have ovarian cancer [2]. Conversely, when a cut-off of 30 IU/ml is used, the test has a sensitivity of 81% and specificity of 75% [15]. In general, one of the methods for assessment of treatment response is by monitoring the CA125 levels. Moreover, longitudinal monitoring of CA125 levels can also provide additional information about survival in ovarian cancer [16].

#### **2.2. Risk of malignancy index (RMI)**

The risk of malignancy index (RMI) is a validated clinical tool that is used to assess the risk of having OC [2]. RMI combines three pre-surgical features which include serum CA125 levels using the unit IU/ml (CA125), menopausal status (M) and ultrasound score (U) for its assessment, using the formula: RMI = CA125 × M × U. The ultrasound result is scored 1 point for each of the following characteristics, namely the presence of multilocular cysts, solid areas, metastases, ascites and bilateral lesions. U = 0 (for an ultrasound score of 0), U = 1 (for an ultrasound score of 1), U = 3 (for an ultrasound score of 2–5). The menopausal status is scored as 1 = pre-menopausal and 3 = post-menopausal. The classification of 'post-menopausal' is a woman who has had no period for more than 1 year or a woman over 50 years of age and has had a hysterectomy [2].

According to the NICE guidelines, RMI scores are interpreted as low, moderate and high risk based on the total score [2]. Low-risk RMI is for scores <25 (noted in 40% of women, and the risk of cancer is <3%). Moderate-risk RMI is assigned for scores 25–250 (noted in 30% of women, and the risk of cancer is 20%), whereas scores >250 are associated with high-risk RMI (observed in 30% of women, and the risk of cancer is 75%). Women with moderate or intermediate risk are recommended to have an MRI for further evaluation of the ovarian lesion. Whereas, post-menopausal women with an RMI score of >250 should be referred to a cancer centre for further assessment and often undergo a staging computed tomography scan.

**Figure 2a** demonstrates a thick-walled, right adnexal lesion and **Figure 2b** demonstrates a

**Figure 2.** Grey scale ultrasound scans demonstrating suspicious bilateral large adnexal masses in a postmenopausal woman. The white arrow on the left indicates the uterus, the white arrow on the right indicates the urinary bladder and the orange arrow indicates the thick-walled right adnexal mass. Image ( b) shows the dimensions of the left adnexal mass.

The Past, Present and Future of Diagnostic Imaging in Ovarian Cancer

http://dx.doi.org/10.5772/intechopen.74449

179

USS has improved specificity in detecting OC by the utility of simple ultrasound rules model [18]. In particular, by using the conventional technique of pattern recognition or subjective assessment by an experienced sonographer, the sensitivity and specificity were 83 and 90%. Whereas, by the technique of using the simple ultrasound rules (**Table 1**) was 92 and 96% respectively [19]. The rules comprised of five ultrasound features to predict a malignant tumour (M features) and five to predict a benign tumour (B features). These include features

**Sonographic characteristics Benign features (B features) Malignant features (M features)**

Papillary projections None/a few, thin Multiple (at least 4), thick

Size of largest solid component None/3–7 mm Usually >7 mm Wall Thin, regular Thick, irregular Internal Doppler flow None/minimal Increased Ascites Absent Present

Acoustic shadow Present Not applicable

**Table 1.** Sonographic characteristics of ovarian lesions based on simple ultrasound rules [18, 19].

Tumour size <100 mm >100 mm Loculations Unilocular, smooth Multilocular Consistency Cystic Solid, mixed

multiseptated, left ovarian lesion in the pouch of Douglas.

**2.4. Interpretation of ultrasound imaging**

#### **2.3. Ultrasound scan**

Ultrasound (USS) is a safe, inexpensive and widely available diagnostic modality. Grey scale USS is a real-time imaging that detects the difference in the acoustic impedance (density × velocity of sound) of internal structures and can give excellent soft tissue detail for the evaluation of adnexal masses. Trans-abdominal scan (TAS) and transvaginal scan (TVS) are the first line diagnostic imaging modality for diagnosing OC. TAS utilises a low frequency (3.5–7 MHz) convex probe to characterise adnexal lesions that have grown beyond the pelvic brim. TVS uses a higher frequency (7.5–12 MHz) endocervical probe and gives better spatial resolution as it is placed closer to the ovaries; and is the first line modality of choice for small masses [16]. Nevertheless, a smaller field of view, leading to a possibility of overlooking a larger pelvic mass, is one of its limitations. Therefore, TAS is usually performed first followed by a TVS as a standard scan procedure.

In women suspected of having ovarian cancer, USS is indicated as a first line diagnostic imaging test. USS can diagnose the presence of an adnexal mass and help characterise it. The size, consistency, presence of loculations and solid component within a tumour; are some of the criteria used to characterise adnexal masses. A point to note is that an anechoic ovarian lesion detected in a postmenopausal woman should be considered as a physiological inclusion cyst, and not a pathological cyst if it was smaller than 10 mm and did not distort the ovary [17]. In certain occasions, the presence of bilateral lesions can be detected, and the pouch of Douglas is a common location for a left ovarian lesion (**Figure 2**).

**Figure 2.** Grey scale ultrasound scans demonstrating suspicious bilateral large adnexal masses in a postmenopausal woman. The white arrow on the left indicates the uterus, the white arrow on the right indicates the urinary bladder and the orange arrow indicates the thick-walled right adnexal mass. Image ( b) shows the dimensions of the left adnexal mass.

**Figure 2a** demonstrates a thick-walled, right adnexal lesion and **Figure 2b** demonstrates a multiseptated, left ovarian lesion in the pouch of Douglas.

#### **2.4. Interpretation of ultrasound imaging**

response is by monitoring the CA125 levels. Moreover, longitudinal monitoring of CA125

The risk of malignancy index (RMI) is a validated clinical tool that is used to assess the risk of having OC [2]. RMI combines three pre-surgical features which include serum CA125 levels using the unit IU/ml (CA125), menopausal status (M) and ultrasound score (U) for its assessment, using the formula: RMI = CA125 × M × U. The ultrasound result is scored 1 point for each of the following characteristics, namely the presence of multilocular cysts, solid areas, metastases, ascites and bilateral lesions. U = 0 (for an ultrasound score of 0), U = 1 (for an ultrasound score of 1), U = 3 (for an ultrasound score of 2–5). The menopausal status is scored as 1 = pre-menopausal and 3 = post-menopausal. The classification of 'post-menopausal' is a woman who has had no period for more than 1 year or a woman over 50 years of age and has had a hysterectomy [2]. According to the NICE guidelines, RMI scores are interpreted as low, moderate and high risk based on the total score [2]. Low-risk RMI is for scores <25 (noted in 40% of women, and the risk of cancer is <3%). Moderate-risk RMI is assigned for scores 25–250 (noted in 30% of women, and the risk of cancer is 20%), whereas scores >250 are associated with high-risk RMI (observed in 30% of women, and the risk of cancer is 75%). Women with moderate or intermediate risk are recommended to have an MRI for further evaluation of the ovarian lesion. Whereas, post-menopausal women with an RMI score of >250 should be referred to a cancer centre for further assessment and often undergo a staging computed tomography scan.

Ultrasound (USS) is a safe, inexpensive and widely available diagnostic modality. Grey scale USS is a real-time imaging that detects the difference in the acoustic impedance (density × velocity of sound) of internal structures and can give excellent soft tissue detail for the evaluation of adnexal masses. Trans-abdominal scan (TAS) and transvaginal scan (TVS) are the first line diagnostic imaging modality for diagnosing OC. TAS utilises a low frequency (3.5–7 MHz) convex probe to characterise adnexal lesions that have grown beyond the pelvic brim. TVS uses a higher frequency (7.5–12 MHz) endocervical probe and gives better spatial resolution as it is placed closer to the ovaries; and is the first line modality of choice for small masses [16]. Nevertheless, a smaller field of view, leading to a possibility of overlooking a larger pelvic mass, is one of its limitations. Therefore, TAS is usually performed first followed

In women suspected of having ovarian cancer, USS is indicated as a first line diagnostic imaging test. USS can diagnose the presence of an adnexal mass and help characterise it. The size, consistency, presence of loculations and solid component within a tumour; are some of the criteria used to characterise adnexal masses. A point to note is that an anechoic ovarian lesion detected in a postmenopausal woman should be considered as a physiological inclusion cyst, and not a pathological cyst if it was smaller than 10 mm and did not distort the ovary [17]. In certain occasions, the presence of bilateral lesions can be detected, and the pouch of Douglas

levels can also provide additional information about survival in ovarian cancer [16].

**2.2. Risk of malignancy index (RMI)**

178 Ovarian Cancer - From Pathogenesis to Treatment

**2.3. Ultrasound scan**

by a TVS as a standard scan procedure.

is a common location for a left ovarian lesion (**Figure 2**).

USS has improved specificity in detecting OC by the utility of simple ultrasound rules model [18]. In particular, by using the conventional technique of pattern recognition or subjective assessment by an experienced sonographer, the sensitivity and specificity were 83 and 90%. Whereas, by the technique of using the simple ultrasound rules (**Table 1**) was 92 and 96% respectively [19]. The rules comprised of five ultrasound features to predict a malignant tumour (M features) and five to predict a benign tumour (B features). These include features


**Table 1.** Sonographic characteristics of ovarian lesions based on simple ultrasound rules [18, 19].

of shape, size, solidity, and results of colour Doppler examination. Masses would be classified as malignant if one or more M features were present in the absence of a B feature. While if one or more B features were present in the absence of an M feature, the mass would be classified as benign. However, if both M features and B features were present, or if none of the features was present, the simple rules were considered inconclusive [20].

Colour Doppler can identify the presence of colour flow, within the papillary or solid components of an ovarian tumour and has good PPV for detecting malignancy. Nevertheless, the absence of colour flow in smaller lesions potentially causes falsely negative observations. False positive findings of flow can also occur in ovarian cysts in the luteal phase in premenopausal women [21].

The sensitivity and specificity of grey scale USS alone has been reported as 88% and 96% respectively; whereas, with the addition of colour Doppler as 83% and 97% respectively [22]. Although the introduction of power 3D Doppler has been able to increase the PPV of detecting malignancy; the availability of instruments and necessary expertise for interpretation has limited the use of this technique [23].

Several scoring systems have been suggested based on USS morphology of ovarian lesions to calculate and determine scores for malignancy [15, 24]. The PPV of these systems are small because the morphology of many benign lesions overlaps with that of malignant disease [21]. Rarely, certain OC are detected in large cysts, usually >7.5 cm in diameter; but do not exhibit an apparent complex morphology on ultrasound [25].

evaluating the potential need to start or change the treatment regime [31]. Contrast-enhanced CT (CECT) studies have an added advantage compared to low dose non-enhanced CT scans, as they enable improved delineation of anatomical structures, and increased sensitivity for detection of pathological lesions [32]. Contrast-enhanced PET/CT is a more accurate imaging

**Figure 3.** Multiplanar computed tomography scan demonstrating an ovarian cancer. The red arrow shows thickened

The Past, Present and Future of Diagnostic Imaging in Ovarian Cancer

http://dx.doi.org/10.5772/intechopen.74449

181

Conventional CT has a limited and variable sensitivity of 40–93% and specificity of 50–98% for detection of recurrent disease [30]. Spiral CT can improve the detection of peritoneal lesions and implants, in particular in those with concurrent ascites. Obtaining a CT before secondary debulking may aid in surgical planning and to assess the feasibility of achieving maximum

Contrast-enhanced CT scans (CECT) can detect the involvement of specific intra-abdominal sites recognised to reduce the chances of optimal debulking. These sites include suprarenal aortic lymph nodes, disease in the root of the mesentery, portal triad disease, or bulky liver disease [35]. Conversely, multidetector CECT scans often underestimates the extent of liver surface disease and infra-renal para-aortic lymph node involvement [36]. The reliability of CT assessment is also related to improvements in imaging techniques as well as scanner equip-

CECT provides improved contrast resolution in delineating suspicious adnexal masses [38]. CECT is helpful in characterising benign and malignant ovarian tumours by observation of certain characteristic features (**Table 2**). It can also help differentiate OC subtypes, albeit with some overlapping features, especially the commoner subtypes such as serous tumours. Serous tumours are usually unilocular but with multiple papillary projections and often present bilaterally [39]. Furthermore, peritoneal carcinomatosis is also seen more frequently in

modality than PET using low dose CT for assessing OC recurrence [33].

peritoneum, black arrow shows gross ascites and white arrow shows a large adnexal mass.

ment and this can vary across different centres [37].

**2.7. Interpretation of computed tomography scans**

serous adenocarcinomas [40].

resectability [34].

#### **2.5. Limitations and future research in ultrasound**

Recently, some experiments have been conducted to evaluate the role of contrast-enhanced USS to help further characterise ovarian tumours [26]. The meta-analysis of 10 studies revealed a pooled sensitivity of 0.89 and specificity of 0.91 respectively [27]. The limitation of ultrasound is the low sensitivity for detection of peritoneal metastasis [28]. Furthermore, screening low-risk population by transvaginal ultrasound may incidentally detect indeterminate lesions and lead to unnecessary biopsies [29]. Therefore, it should not be considered as a standalone investigation to be used to screen the general population for OC.

#### **2.6. Computed tomography (CT) scan**

Computed tomography (CT) scan utilises ionising radiation (photon beams of X-ray) to create cross sectional images of the internal organs. CT scans can give detailed information regarding tumour extent and metastatic disease (**Figure 3**). **Figure 3** is a multiplanar CT scan of a patient in axial, coronal and sagittal views, demonstrating a large ovarian cancer with intraabdominal extension (white arrow) as well as gross ascites (black arrow) and thickened peritoneum consistent with metastasis (red arrow).

It is the preferred modality for the staging of OC and detection of recurrence because it is more widely available and less costly compared to magnetic resonance imaging (MRI) [30]. The Response Evaluation Criteria in Solid Tumours (RECIST) is often used in assessing treatment response in follow up CT scans and may be employed alone or in combination with CA125, for

**Figure 3.** Multiplanar computed tomography scan demonstrating an ovarian cancer. The red arrow shows thickened peritoneum, black arrow shows gross ascites and white arrow shows a large adnexal mass.

evaluating the potential need to start or change the treatment regime [31]. Contrast-enhanced CT (CECT) studies have an added advantage compared to low dose non-enhanced CT scans, as they enable improved delineation of anatomical structures, and increased sensitivity for detection of pathological lesions [32]. Contrast-enhanced PET/CT is a more accurate imaging modality than PET using low dose CT for assessing OC recurrence [33].

Conventional CT has a limited and variable sensitivity of 40–93% and specificity of 50–98% for detection of recurrent disease [30]. Spiral CT can improve the detection of peritoneal lesions and implants, in particular in those with concurrent ascites. Obtaining a CT before secondary debulking may aid in surgical planning and to assess the feasibility of achieving maximum resectability [34].

Contrast-enhanced CT scans (CECT) can detect the involvement of specific intra-abdominal sites recognised to reduce the chances of optimal debulking. These sites include suprarenal aortic lymph nodes, disease in the root of the mesentery, portal triad disease, or bulky liver disease [35]. Conversely, multidetector CECT scans often underestimates the extent of liver surface disease and infra-renal para-aortic lymph node involvement [36]. The reliability of CT assessment is also related to improvements in imaging techniques as well as scanner equipment and this can vary across different centres [37].

#### **2.7. Interpretation of computed tomography scans**

of shape, size, solidity, and results of colour Doppler examination. Masses would be classified as malignant if one or more M features were present in the absence of a B feature. While if one or more B features were present in the absence of an M feature, the mass would be classified as benign. However, if both M features and B features were present, or if none of the features

Colour Doppler can identify the presence of colour flow, within the papillary or solid components of an ovarian tumour and has good PPV for detecting malignancy. Nevertheless, the absence of colour flow in smaller lesions potentially causes falsely negative observations. False positive findings of flow can also occur in ovarian cysts in the luteal phase in premeno-

The sensitivity and specificity of grey scale USS alone has been reported as 88% and 96% respectively; whereas, with the addition of colour Doppler as 83% and 97% respectively [22]. Although the introduction of power 3D Doppler has been able to increase the PPV of detecting malignancy; the availability of instruments and necessary expertise for interpretation has

Several scoring systems have been suggested based on USS morphology of ovarian lesions to calculate and determine scores for malignancy [15, 24]. The PPV of these systems are small because the morphology of many benign lesions overlaps with that of malignant disease [21]. Rarely, certain OC are detected in large cysts, usually >7.5 cm in diameter; but do not exhibit

Recently, some experiments have been conducted to evaluate the role of contrast-enhanced USS to help further characterise ovarian tumours [26]. The meta-analysis of 10 studies revealed a pooled sensitivity of 0.89 and specificity of 0.91 respectively [27]. The limitation of ultrasound is the low sensitivity for detection of peritoneal metastasis [28]. Furthermore, screening low-risk population by transvaginal ultrasound may incidentally detect indeterminate lesions and lead to unnecessary biopsies [29]. Therefore, it should not be considered as a

Computed tomography (CT) scan utilises ionising radiation (photon beams of X-ray) to create cross sectional images of the internal organs. CT scans can give detailed information regarding tumour extent and metastatic disease (**Figure 3**). **Figure 3** is a multiplanar CT scan of a patient in axial, coronal and sagittal views, demonstrating a large ovarian cancer with intraabdominal extension (white arrow) as well as gross ascites (black arrow) and thickened peri-

It is the preferred modality for the staging of OC and detection of recurrence because it is more widely available and less costly compared to magnetic resonance imaging (MRI) [30]. The Response Evaluation Criteria in Solid Tumours (RECIST) is often used in assessing treatment response in follow up CT scans and may be employed alone or in combination with CA125, for

standalone investigation to be used to screen the general population for OC.

was present, the simple rules were considered inconclusive [20].

pausal women [21].

limited the use of this technique [23].

180 Ovarian Cancer - From Pathogenesis to Treatment

**2.6. Computed tomography (CT) scan**

toneum consistent with metastasis (red arrow).

an apparent complex morphology on ultrasound [25].

**2.5. Limitations and future research in ultrasound**

CECT provides improved contrast resolution in delineating suspicious adnexal masses [38]. CECT is helpful in characterising benign and malignant ovarian tumours by observation of certain characteristic features (**Table 2**). It can also help differentiate OC subtypes, albeit with some overlapping features, especially the commoner subtypes such as serous tumours. Serous tumours are usually unilocular but with multiple papillary projections and often present bilaterally [39]. Furthermore, peritoneal carcinomatosis is also seen more frequently in serous adenocarcinomas [40].


identify the fatty tissue components within individual adnexal tumours. It can delineate ovarian lesions from uterine or urinary bladder involvement. MRI is also useful in cases where CECT is relatively contraindicated such as in the pregnant woman, in a patient with, a history of dye allergy or where giving iodinated contrast material is contraindicated, e.g., in renal

The Past, Present and Future of Diagnostic Imaging in Ovarian Cancer

http://dx.doi.org/10.5772/intechopen.74449

183

MRI is able to differentiate simple ovarian cysts from malignant lesions with solid internal components. Simple cysts return a low signal in the case of T1-weighted images and a high signal in T2-weighted images, whereas malignant lesions are often heterogeneous and show marked enhancement of its solid components. MRI is best to delineate the local extent of the

The utility of whole-body diffusion-weighted imaging in magnetic resonance imaging (WB-DWI/MRI) has shown some promising results. Diffusion-weighted imaging measures the Brownian motion of extracellular water and thereby approximates tissue cellularity and fluid viscosity, hence malignant tumours that have increased cellularity will have restricted diffusion, thus giving lower apparent diffusion coefficient (ADC) values. Interestingly, DWI/ ADC sequences have shown 94% accuracy for primary tumour characterisation which is comparable with the results of 18F-Fluorodeoxyglucose positron emission tomography/com-

WB-DWI/MRI has also shown an improved accuracy of 91% for peritoneal staging compared with CT (75%) and 18F-FDG PET/CT (71%). It also has higher accuracy (87%) for detecting retroperitoneal lymphadenopathies compared to CT scans (71%) [42]. However, this study was limited by the relatively small number of cases, the discrepancy between the ratio of MRI and PET/CT cases performed as well as the relatively large number of patients who presented with advanced disease, potentially increasing the pre-test likelihood of detecting

Molecular imaging, namely Positron Emission Tomography-Computed Tomography (PET/ CT) scans are indicated for the detection of recurrence of OC as even small volume disease can easily be detected. It is considered by the European Society of Medical Oncology (ESMO) as an appropriate imaging modality to help in the selection of patients for secondary debulking surgery. PET/CT can alter the management plan in metastatic ovarian cancers by detecting additional sites of disease not seen on CT scans, and identifying locations that are not ame-

PET/CT can assess tumour aggressiveness by demonstrating an elevated level of the injected radiopharmaceutical, e.g., 18F-Fluorodeoxyglucose (18F-FDG) that is trapped in tumour cells, as quantified by standardised uptake values (SUV) [43]. Interestingly, elevated maximum

**2.12. Positron emission tomography-computed tomography (PET/CT) scan**

impairment. Hence, a non-contrast-enhanced MRI scan should be performed instead.

**2.10. Interpretation of magnetic resonance imaging**

tumour as well as detect pelvic nodal metastases.

puted tomography (18F-FDG PET/CT) [42].

metastases.

nable to cytoreduction [1].

**2.11. Limitations and future research in magnetic resonance imaging**

**Table 2.** Computed tomography characteristics of adnexal masses (adapted from Suppiah et al. [38] and Jung et al. [9]).

#### **2.8. Limitations and future research in computed tomography**

The main limitation of a CT scan is its inability to detect deposits on bowel serosa, mesentery and omental regions that are smaller than 5 mm; especially in the absence of ascites. However, this can be solved by pre-surgical laparoscopic assessment. The detection of subdiaphragmatic peritoneal deposits are also difficult, but can be aided by multiplanar reformatting of contrast-enhanced scans.

#### **2.9. Magnetic resonance imaging (MRI)**

Magnetic resonance imaging (MRI) uses a high strength magnetic field and pulsed radiofrequency waves to generate images with excellent soft tissue detail. It does not utilise ionising radiation and is relatively safe to use. It commonly involves acquiring T1-weighted, T2-weighted, fat-saturation spin echo, and usually, includes post-gadolinium contrastenhanced T1-weighted fat-saturation sequences for the pelvic region. This protocol includes a full abdominal scan in three planes for the staging of ovarian cancer [41].

The ability of MRI to correctly stage ovarian cancer is excellent, providing a sensitivity and specificity of 98 and 88% respectively, as compared to 92 and 89% respectively for CECT [41]. MRI and CT have been noted to be more sensitive than ultrasound for detection of peritoneal metastases. The accuracy of MRI (76%) is also better than CT (57%) for detection of lymph nodes [28]. The improved soft tissue resolution achieved by MRI is able to better delineate the presence of pathological lymph nodes, both within the pelvic cavity as well as extra-pelvic spread. However, its limitations are that it is rather costly, time-consuming and often difficult to interpret due to breathing and bowel movement artefacts.

Therefore, the clinical utility of MRI is limited to evaluation of indeterminate pelvic lesions. MRI can detect haemorrhagic lesions and enhancement in papillary projections, as well as identify the fatty tissue components within individual adnexal tumours. It can delineate ovarian lesions from uterine or urinary bladder involvement. MRI is also useful in cases where CECT is relatively contraindicated such as in the pregnant woman, in a patient with, a history of dye allergy or where giving iodinated contrast material is contraindicated, e.g., in renal impairment. Hence, a non-contrast-enhanced MRI scan should be performed instead.

#### **2.10. Interpretation of magnetic resonance imaging**

**2.8. Limitations and future research in computed tomography**

**CT scan characteristics Benign Malignant** Size <4 cm >4 cm Consistency Cystic Mixed/Solid Papillary projections Absent/a few Multiple Wall Thin, regular Thick, irregular Internal calcifications Occasionally present Infrequently present

Pelvic lymphadenopathy Absent Present Laterality Unilateral Bilateral Ascites Absent Present Peritoneal involvement Absent Present Distant organ metastasis Absent Present

matting of contrast-enhanced scans.

182 Ovarian Cancer - From Pathogenesis to Treatment

**2.9. Magnetic resonance imaging (MRI)**

The main limitation of a CT scan is its inability to detect deposits on bowel serosa, mesentery and omental regions that are smaller than 5 mm; especially in the absence of ascites. However, this can be solved by pre-surgical laparoscopic assessment. The detection of subdiaphragmatic peritoneal deposits are also difficult, but can be aided by multiplanar refor-

**Table 2.** Computed tomography characteristics of adnexal masses (adapted from Suppiah et al. [38] and Jung et al. [9]).

Magnetic resonance imaging (MRI) uses a high strength magnetic field and pulsed radiofrequency waves to generate images with excellent soft tissue detail. It does not utilise ionising radiation and is relatively safe to use. It commonly involves acquiring T1-weighted, T2-weighted, fat-saturation spin echo, and usually, includes post-gadolinium contrastenhanced T1-weighted fat-saturation sequences for the pelvic region. This protocol includes a

The ability of MRI to correctly stage ovarian cancer is excellent, providing a sensitivity and specificity of 98 and 88% respectively, as compared to 92 and 89% respectively for CECT [41]. MRI and CT have been noted to be more sensitive than ultrasound for detection of peritoneal metastases. The accuracy of MRI (76%) is also better than CT (57%) for detection of lymph nodes [28]. The improved soft tissue resolution achieved by MRI is able to better delineate the presence of pathological lymph nodes, both within the pelvic cavity as well as extra-pelvic spread. However, its limitations are that it is rather costly, time-consuming and often difficult

Therefore, the clinical utility of MRI is limited to evaluation of indeterminate pelvic lesions. MRI can detect haemorrhagic lesions and enhancement in papillary projections, as well as

full abdominal scan in three planes for the staging of ovarian cancer [41].

to interpret due to breathing and bowel movement artefacts.

MRI is able to differentiate simple ovarian cysts from malignant lesions with solid internal components. Simple cysts return a low signal in the case of T1-weighted images and a high signal in T2-weighted images, whereas malignant lesions are often heterogeneous and show marked enhancement of its solid components. MRI is best to delineate the local extent of the tumour as well as detect pelvic nodal metastases.

#### **2.11. Limitations and future research in magnetic resonance imaging**

The utility of whole-body diffusion-weighted imaging in magnetic resonance imaging (WB-DWI/MRI) has shown some promising results. Diffusion-weighted imaging measures the Brownian motion of extracellular water and thereby approximates tissue cellularity and fluid viscosity, hence malignant tumours that have increased cellularity will have restricted diffusion, thus giving lower apparent diffusion coefficient (ADC) values. Interestingly, DWI/ ADC sequences have shown 94% accuracy for primary tumour characterisation which is comparable with the results of 18F-Fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) [42].

WB-DWI/MRI has also shown an improved accuracy of 91% for peritoneal staging compared with CT (75%) and 18F-FDG PET/CT (71%). It also has higher accuracy (87%) for detecting retroperitoneal lymphadenopathies compared to CT scans (71%) [42]. However, this study was limited by the relatively small number of cases, the discrepancy between the ratio of MRI and PET/CT cases performed as well as the relatively large number of patients who presented with advanced disease, potentially increasing the pre-test likelihood of detecting metastases.

#### **2.12. Positron emission tomography-computed tomography (PET/CT) scan**

Molecular imaging, namely Positron Emission Tomography-Computed Tomography (PET/ CT) scans are indicated for the detection of recurrence of OC as even small volume disease can easily be detected. It is considered by the European Society of Medical Oncology (ESMO) as an appropriate imaging modality to help in the selection of patients for secondary debulking surgery. PET/CT can alter the management plan in metastatic ovarian cancers by detecting additional sites of disease not seen on CT scans, and identifying locations that are not amenable to cytoreduction [1].

PET/CT can assess tumour aggressiveness by demonstrating an elevated level of the injected radiopharmaceutical, e.g., 18F-Fluorodeoxyglucose (18F-FDG) that is trapped in tumour cells, as quantified by standardised uptake values (SUV) [43]. Interestingly, elevated maximum standardised uptake values (SUVmax) are frequently detected in the ovaries in the luteal phase of the menstrual cycle. This is considered as normal physiological FDG metabolism and should not be mistaken for pathology. Therefore, PET/CT scans should be scheduled right after the menstruation to minimise this observed effect in premenopausal women.

PET/CT scans are performed using a hybrid PET/CT scanner commonly using 3D lutetium oxy-orthosilicate crystals as detectors for the PET component. It is recommended that the examination include a diagnostic contrast-enhanced computed tomography (CECT) scan by the administration of low osmolar iodinated contrast media. Apart from enabling attenuation correction, and anatomical localisation; CECT is essential for performing diagnostic clinical staging [1]. Patients are instructed to fast for a minimum of 6 h before scanning, and blood glucose is checked before the scan. Subsequently, 18F-FDG will be administered and subjects are kept in a dark room for approximately 60 min to allow for uptake time. Subjects are given approximately 100 mL (2 ml/kg body weight) of iodinated contrast media during the CECT scan. Immediately after the CT, PET images acquisition will be performed over the same anatomic regions. The attenuation corrected CECT images will then be fused with PET images. The combined images will be utilised for visual interpretation, tumour size and maximum standard uptake value (SUVmax) measurements.
