JNM: PERCIST tops RECIST for esophageal cancer eval
18F-FDG PET/CT image of 58-year-old man with cancer in lower thoracic esophagus.
Source: J Nucl Med 2012;53(6):872-880
PET response criteria in solid tumors (PERCIST) provided an independent predictor of outcomes among patients with esophageal cancer undergoing neoadjuvant chemotherapy, according to a study published in the June issue of the Journal of Nuclear Medicine.

Although the Response Evaluation Criteria in Solid Tumors (RECIST) have been widely employed as an anatomic tumor response metric, the system is limited. Tumors with obscure margins can be difficult to measure, and molecularly targeted treatments do not necessarily produce changes in morphologic size as measured by RECIST. Thus, PERCIST, which measures chemotherapeutic response metabolically and quantitatively, has been proposed as a standardized method to quantify metabolic tumor response.

Pathologic venous invasion and resection level offered significant prognostic factors for predicting survival or recurrence, according to Masahiro Yanagawa, MD, from the department of radiology at Osaka University Graduate School of Medicine in Japan, and colleagues.

While neither measure is readily obtainable prior to surgery, PERCIST offers a noninvasive measure of treatment response.

The researchers devised a retrospective analysis to evaluate PERCIST among patients with esophageal cancer undergoing neoadjuvant chemotherapy. The study included 51 patients who underwent neoadjuvant chemotherapy and surgery between June 2007 and March 2010.

The researchers leveraged peak SUV on PET exams to measure complete metabolic response and partial metabolic response. Yanagawa and colleagues used PERCIST to evaluate the hottest lesion and measured the anatomic size change of the single primary site in PERCIST and longest diameter according to RECIST.

They developed a five-point ordinal scale for both RECIST and PERCIST to compare the treatment response assessments and reported a significant difference in the results of response classification between the two measures.

According to RECIST, the distribution of pathologic responses was: 13 grade 1a, seven grade 1b, four grade 2 and two grade 3. PERCIST resulted in 23 grade 1a, 11 grade 1b, six grade 2 and three grade 3. In both systems, good chemotherapeutic responses did not always correlate with good pathologic responses.

Reduction of tumor diameter on CT and reduction of SUVpeak normalized to body mass did not predict survival and recurrence. But PERCIST better correlated with prognostic measures; the authors suggested that this may be because it included reduction of SUVpeak normalized to body mass and its value after chemotherapy.  

The researchers emphasized the importance of prognostic data, as they can inform patient selection for multidisciplinary treatments and may allow some patients to forego surgery based on chemotherapy response.

Yanagawa et al noted that additional studies are required to determine how to optimally employ PERCIST but predicted that the tool will provide improved information for therapeutic selection among cancer patients.