The presurgical modified Glasgow prognostic score can predict oncologic outcomes in patients with nonmetastatic upper tract urothelial carcinoma (UTUC), recent studies suggest.

In a study published in Urologic Oncology, 2492 patients scheduled for radical nephroureterectomy (RNU) obtained their modified Glasgow prognostic score (mGPS), a combination of preoperative levels of C-reactive protein (CRP) and albumin. Of the cohort, 1929 (77%), 530 (21%), and 33 (1%) had an mGPS of 0, 1, and 2, respectively. Patients with a CRP level greater than 0.5 mg/dL in addition to hypoalbuminemia received a score of 2, those with only 1 of these factors received a score of 1, and those with normal serum CRP and albumin levels received a score of 0.

The mGPS independently predicted lymph node invasion (LNI) and nonorgan-confined disease (NOC) at RNU, Shahrokh F. Shariat, MD, of the Medical University of Vienna in Austria, and colleagues reported. Compared with mGPS 0, mGPS 1 was associated with 1.8- and 1.3-fold greater odds for LNI and NOC , respectively (P <.001 and P =008, respectively); mGPS 2 was significantly associated with 2.8-fold greater odds for LNI (P =.02) and nonsignificant 1.5-fold greater odds for NOC disease. Higher mGPS also was generally associated with higher tumor stage and grade, positive lymph node status, necrosis, and sessile architecture.

On multivariable analysis, preoperative mGPS 1 was significantly associated with an approximately 1.7-, 1.8-, and 1.5-fold increased risk for recurrence, cancer-specific death, and death from any cause, respectively, compared with preoperative mGPS 0 (all P <.001). Results showed that mGPS 2 was not significantly associated with these outcomes.

In the post-operative setting, mGPS 1 was significantly associated with an approximately 1.3-, 1.4-, and 1.2-fold increased risk for recurrence, cancer-specific death, and death from any cause, respectively, compared with mGPS 0 (P =.008, P =.002, and P =.039). Again, compared with mGPS 0, mGPS 2 was not significantly associated with these outcomes.

Incorporating mGPS improved the discrimination of a preoperative model for predicting oncologic outcomes compared with standard prognosticators, the investigators reported.
According to Dr Shariat’s team, “preoperative prediction of advanced disease at RNU and of worse survival outcomes is essential to select patients for neoadjuvant systemic therapies such as chemotherapy or immunotherapy and, ideally, to refine the indication and extent of lymph node dissection.”

A second study of 273 patients with UTUC undergoing RNU, which was published in the Japanese Journal of Clinical Oncology, found that mGPS 2 was independently associated with worse RFS, CSS, and OS.

Of the 273 patients, 216 (79%), 5 (17%), and 12 (4%) had an mGPS of 0, 1, and 2, respectively. During a median follow-up period of 36.1 months, there were 86 recurrences, 49 cancer-specific deaths, and 70 deaths due to any causes, Shunsuke Tsuzuki, MD, of Jikei University School of Medicine in Tokyo, Japan, and colleagues reported. The 5-year RFS rates for patients with mGPS 0, 1 and 2 were 49.4%, 23.8% and 10.6%, respectively (0 vs 2: P <.01). The 5-year CSS rates for patients with mGPS 0, 1 and 2 were 80.4%, 72.9% and 32.3%, respectively, with a significantly higher rate associated with mGPS 2 compared with mGPS 0 (0 vs 2: P <.01). The 5-year OS rates for patients with mGPS 0, 1 and 2 were 74.4%, 57.3%, and 32.3%, respectively. Compared with mGPS 0, only mGPS 2 was significantly associated with worse survival (P <.01).

“High pre-surgical mGPS was significantly associated with mortality of UTUC patients treated with RNU,” the authors concluded. “Therefore, assessment of presurgical mGPS could help predict the patients who will meet disease recurrence and threaten the survival.”

Disclosure: Several study authors declared affiliations with the pharmaceutical industry. Please see the original references for a full list of authors’ disclosures.