Effects of Preoperative Serum Prolactin and Tumor Volume on Postoperative Remission in a Cohort of 219 Prolactinoma Resections

Objective: Prolactinomas can be managed medically with surgery reserved as second-line therapy. Postoperative remission can be difficult to predict, and prior studies have not examined how volume-adjusted prolactin relates to postoperative outcomes. We assessed the correlation between presurgical tumor volume and prolactin as well as its effect on surgical outcomes.

Methods: A total of 219 patients undergoing transsphenoidal prolactinoma resection from 2012 to 2019 were reviewed. Cases were compared between patients with and without biochemical remission. Linear regression examined associations between tumor volume and serum prolactin. Volume-adjusted prolactin was defined as serum prolactin divided by tumor volume. Receiver operating characteristic curves (ROC) were generated to determine thresholds that predicted remission status.

Results: Linear regression (p < 0.001) revealed an association between pre-op tumor volume and prolactin; women demonstrated a greater rise in prolactin per volume increase (185.70 vs. 79.77 μg/L/cm³; p < 0.001). Patients achieving remission exhibited a 66.08 ug/L increase in preoperative prolactin/cm³ of preoperative tumor volume (p < 0.001, [Fig. 1A]), lower than the 111.46 ug/L/cm³ in patients without remission (p < 0.001, [Fig. 1B]). ROC curves revealed thresholds that predicted remission failure ([Table 1]). Stratified by degree of cavernous sinus invasion, thresholds were found for both serum prolactin (no invasion: 113.5 μg/L; invasion: 282.0 μg/L; Knosp <3: 345.95 μg/L) and tumor volume (invasion: 4.85 cm³; Knosp <3: 4.18 cm³). Thresholds were rendered nonsignificant in patients with Knosp grade ≥ 3.

Conclusions: While we found a correlation between prolactinoma volume and serum prolactin levels among all patients, those without remission had a greater rise in serum prolactin/cm³ of preoperative tumor volume than those achieving remission, suggesting unique a tumor composition. We also identified preoperative prolactin and tumor volume thresholds where biochemical remission status was optimally predicted. These results will help identify prolactinomas where surgery offers an increased or decreased likelihood of postoperative remission.

Fig. 1 Prolactin rise per volume increase, by remission status.

Table 1 ROC analysis predicting biochemical remission failure

Variable	Threshold	Sensitivity	Specificity	AUC	p-Value
No cavernous sinus invasion
Pre-op Prolactin (µg/L)	113.50	0.640	0.792	0.730	<0.001
Pre-op volume (cm3)	4.71	0.960	0.333	0.581	0.305
Cavernous Sinus Invasion
Pre-op prolactin (µg/L)	282.00	0.650	0.851	0.717	0.003
Pre-op volume (cm3)	4.85	0.750	0.702	0.698	0.006
Knosp < 3
Pre-op prolactin (µg/L)	345.95	0.935	0.548	0.779	<0.001
Pre-op volume (cm3)	4.18	0.928	0.429	0.672	0.001
Knosp ≥ 3
Pre-op prolactin (µg/L)	302.85	0.429	0.897	0.606	0.413
Pre-op volume (cm3)	4.48	0.429	0.897	0.596	0.491

Publication History

Article published online:
01 February 2023