low pelvic cancers with a distal margin threatening the preservation of the anal canal or anterior compartment structures (prostate/seminal vesicles in men; vagina and uterus in women) endorectal ultrasound (ERUS), when performed by experienced practitioners, can provide complementary information to assist decision‐making. The small volume of mesorectal fat between the distal rectum and anterior mesorectal margin can limit accuracy of MRI for detecting transmural extension of the tumor and anterior margin involvement by low rectal cancers.
Several retrospective studies correlating preoperative MRI staging of the circumferential resection margin (CRM) with histological CRM involvement have demonstrated decreased reliability of MRI for patients with low, anterior rectal cancer, with MRI mostly overestimating CRM involvement [25, 26]. In the authors’ experience, if MRI shows convincing involvement of anterior structures by cancer, total pelvic exenteration or posterior exenteration is recommended, to achieve R0 resection. However, where involvement of the anterior margin is equivocal, we recommend ERUS when feasible [27, 28]. In one study including 32 patients with anterior rectal cancers, MRI and ERUS had equivalent positive and negative predictive values of 66.6 and 95.6% respectively [27]. A recent retrospective study of 24 patients observed that use of ERUS as an adjunct to MRI improved diagnostic accuracy for anterior margin involvement [29]. ERUS can also provide an accurate assessment of distal cancer margin in relation to the puborectalis muscle when considering intersphincteric dissection and preservation of the anal canal. However, clinically reliable ERUS requires experience of both endoluminal ultrasound and rigid sigmoidoscopy to optimize probe position and scan interpretation [30].
PET‐CT provides complementary information on tumor function and activity [31]. PET‐CT is frequently used in complex cancer patients as a complementary test to CT and liver MRI, particularly for exclusion of metastatic disease in uncommon sites or to help troubleshoot uncertain imaging findings, for example helping distinguish tumor from scar tissue or presence of nodal involvement. PET‐CT utility is limited by poor spatial resolution, anatomical mismatch between sites of disease and displayed metabolic activity (due to patient movement or bowel peristalsis), and false positives generated by sites of inflammation or tissue healing [32].
Radiological Assessment of Cancer Anatomy by MRI
T2‐weighted MRI is the reference standard for assessment of tumor anatomy and resectability. There are two main radiological approaches for interpretation and reporting.
The first is tumor categorization according to the pelvic compartments affected, which can help determine patient prognosis. Various tumor categorization systems have been proposed and are summarized in Table 3.1. The Mayo Clinic classification is based on the presence of symptoms and the number of sites of fixation of the tumor to surrounding pelvic structures [33]; the Yamada classification describes broad categories of localized, sacral, and lateral fixation [34]; and the Wanebo classification is based on the UICC TNM system distinguishing bony or ligamentous pelvic involvement from non‐bony fixation [35]. The Memorial Sloan Kettering classification distinguishes four pelvic compartments [36], while the Royal Marsden Hospital classification distinguishes seven compartments [37]. Some of these systems have attempted to prognosticate as well, although any associations with survival outcomes must be interpreted with caution, as the different institutions had different patient populations and markedly variable R0 resection rates. Moreover, prognostication based on a compartment‐based classification is inherently subject to institutional views of which cancers are resectable according to availability of expertise at that time and not necessarily considering newer surgical techniques [6, 9, 11]. The emergence and implementation of these new techniques has changed opinion on which cancers are resectable and shifted the emphasis away from relying on compartment‐based radiological assessment alone. Indeed, most compartment‐based approaches were designed and validated retrospectively, i.e. after exenterative surgery had already been performed [37].
Table 3.1 Existing classification systems for pelvic compartments.
| Group | Criteria for classification | Definitions | |
|---|---|---|---|
| Mayo Clinic | Symptoms | S0 | Asymptomatic |
| S1 | Symptomatic without pain | ||
| S2 | Symptomatic with pain | ||
| Tumor fixation | F0 | No fixation | |
| F1 | Fixation to one point | ||
| F2 | Fixation to two points | ||
| F3 | Fixation to more than two points | ||
| Yamada | Pattern of pelvic fixation | Localized | Invasion to adjacent pelvic organs/tissues |
| Sacral invasive | Invasion to lower sacrum (≥ S3), coccyx, periosteum | ||
| Lateral invasive | Invasion to sciatic nerve, greater sciatic notch, pelvic sidewall, upper sacrum (S1/2) | ||
| Wanebo | Stages | TR1 | Limited invasion of muscularis |
| TR2 | Full thickness invasion of muscularis propria | ||
| TR3 | Anastomotic recurrence penetrating beyond bowel wall into perirectal soft tissue | ||
| TR4 | Invasion into adjacent organs without fixation | ||
| TR5 | Invasion of bony/ligamentous pelvis | ||
| Memorial Sloan Kettering | Anatomic region | Axial | Anastomotic, mesorectal, perirectal soft tissue, perineum |
| Anterior | Genitourinary tract | ||
| Posterior | Sacrum and presacral fascia | ||
| Lateral | Soft tissues of the pelvic sidewall and lateral bony pelvis | ||
| Royal Marsden Hospital | Planes of dissection on MRI | Central | (Neo)rectum Intraluminal recurrence Perirectal fat or mesorectal, extraluminal recurrence |
| PR | Rectovesical pouch or recto‐uterine pouch of Douglas | ||
| AA PR | Ureters and iliac vessels above peritoneal reflection Sigmoid colon Small bowel Lateral sidewall fascia | ||
| AB PR | Genito‐urinary tract | ||
|
|
