Premium
Residual pulmonary masses after chemotherapy for metastatic nonseminomatous germ cell tumor
Author(s) -
Steyerberg Ewout W.,
Keizer H. Jan,
Messemer Jonathan E.,
Toner Guy C.,
Koops Heimen Schraffordt,
Fosså Sophie D.,
Gerl Arthur,
Sleijfer Dirk T.,
Donohue John P.,
Habbema J. Dik F.
Publication year - 1997
Publication title -
cancer
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.052
H-Index - 304
eISSN - 1097-0142
pISSN - 0008-543X
DOI - 10.1002/(sici)1097-0142(19970115)79:2<345::aid-cncr18>3.0.co;2-0
Subject(s) - medicine , chemotherapy , germ cell , oncology , cancer research , biochemistry , chemistry , gene
BACKGROUND After chemotherapy for a metastatic nonseminomatous germ cell tumor, pulmonary masses may be seen on a computed tomography scan. These residual masses may contain one of three histologic elements: necrosis, mature teratoma, or cancer. Because surgical resection of masses containing only necrosis is unnecessary, the authors aimed to predict the histology of these residual masses. METHODS Six study groups contributed patient data on a total of 215 patients undergoing thoracotomy after cisplatin‐based induction chemotherapy for metastatic testicular nonseminomatous germ cell tumors. Logistic regression analysis was used to estimate the probability of necrosis, mature teratoma, and cancer in relation to predictors known before thoracotomy. RESULTS The pulmonary mass histology was necrosis in 116 patients (54%), mature teratoma in 70 (33%), and cancer in 29 (13%). Necrosis was found at thoracotomy in 89% of those patients with necrosis at retroperitoneal lymph node dissection (RPLND). Other predictors included the primary tumor histology, prechemotherapy tumor marker levels, change in mass size during chemotherapy, and the presence of a single, unilateral mass. Multivariate combination of predictors yielded reliable models (goodness‐of‐fit tests, P > 0.20), which discriminated necrosis well from other histologies, especially if RPLND histology was available (area under the receiver operating characteristic curve, 0.86). CONCLUSIONS This analysis indicated subgroups of patients with a high probability of necrosis and a low risk of cancer for whom close follow‐up of the residual pulmonary mass might be considered. In most patients, a RPLND should be performed before a thoracotomy is considered, because the probability of necrosis is generally higher at thoracotomy than at RPLND and the histology at RPLND is a strong predictor of the histology at thoracotomy. Cancer 1997; 79:345‐55. © 1997 American Cancer Society.