A man who is suspected of having prostate cancer, or whose cancer has been biopsy confirmed, can develop the best right treatment plan if he also has a multiparametric MRI (mpMRI) of the prostate. When his own doctor sends him to an imaging center, his scan is “read” by the center’s radiologist who writes a report. The report consists of technical and clinical terms used by the reader to interpret the images. The report, along with a CD of the images, is sent back to the patient’s doctor.

However, not all radiologists use exactly the same terms to convey their diagnosis. Here is an actual example from a report (I was not the radiologist who wrote it):

Although the central gland is distorted with prominent nodules which are well marginated of which most show high signal intensity, there is a solitary nodule within the right mid-gland, laterally which abuts the capsule and is of low signal intensity (image 16)…In the peripheral zone on the left side near the apex, there is distortion of the capsule associated with mass effect in this region, but there is a low index of suspicion for any extracapsular extension. Due to the mass effect and low signal intensity in this region (image 20), there is a high index of suspicion for moderate-to-high grade adenocarcinoma.

Suppose a different radiologist interpreted the same images. Would that person use identical terms? There would be a great deal of similarity, assuming equal experience on the part of the readers. This particular report mentions a “low index of suspicion for any extracapsular extension.”  This is really important information! A patient whose MRI indicates extracapsular extension (ECE, or tumor that has penetrated the outer margin of the gland indicating early local cancer spread) might not be a candidate for prostatectomy. However, a recent study on mpMRI found that readers use as many as 38 different terms to report ECE.[i] Assuming that these reports are sent to urologists, who have far less (if any) experience reading MRIs, and who may not understand what each individual parameter is assessing, the urologist may be uncertain of the radiologist’s diagnosis.

To solve that problem, enormous strides have been made to come to professional agreement on a standardized “dictionary” of terms that are organized along a 5-point scale from least likely to be cancer (value of 1) to most likely (value of 5). This is called a 5-point standardized diagnostic certainty lexicon. (A lexicon in this sense is a dedicated vocabulary.) A good example already in use is the PI-RADS scale ( see https://sperlingprostatecenter.com/pi-rads-score/).

The new study mentioned above is from a team at Memorial Sloan-Kettering Cancer Center. They compared the reports of radiologists who, without the use of a 5-point lexicon, interpreted the presence/absence of ECE for 254 MRIs of men who were schedule for prostatectomy. The radiologists were then trained on a 5-point lexicon, which they used to interpret 211 more MRIs of men going into prostatectomy. All interpretation results were compared with the actual prostate specimens removed during surgery. Those that used the predefined terms/rating system had the same accuracy rates of identifying ECE as the best MRI, but the language/rating scale were clearer communicators of the diagnosis of ECE. The authors concluded: “The implementation of a lexicon of diagnostic certainty dramatically reduced the number of expressions used by radiologists to indicate their levels of diagnostic certainty… Thus, the use of such a lexicon might prevent miscommunication and help referring clinicians reliably incorporate radiologists’ assessments into clinical decision making.”

For radiologists, the use of a 5-point standardized lexicon makes it more possible for their reports to be fully understood by the urologist who is caring for a prostate cancer patient. A diagnosis of ECE (or no evidence of ECE) will greatly influence the treatment options that are open to that patient. The day is coming when we’ll all speak the same language, which will facilitate improved teamwork between radiology and urology.

 

---

[i]  Wibmer A, Vargas HA, Sosa R et al. Value of a standardized lexicon for reporting levels of diagnostic certainty in prostate MRI. AJR Am J Roentgenol. 2014 Dec;203(6):W651-7.