Correspondence
Use of MRI to Detect Lymph-Node Metastases in Prostate Cancer
N Engl J Med 2003; 349:1185-1186September 18, 2003
- Article
To the Editor:
The results reported by Harisinghani et al. (June 19 issue)1 may have been influenced by the limited dissection of lymph nodes in their study. In most patients, only obturator nodes were dissected, yielding 334 nodes from 80 patients — an average of 4.2 nodes per patient. In contrast, one study of an extended node dissection in a similar group of patients with prostate cancer2 found a median of 21 nodes per patient in the histologic specimens. Of the patients with positive nodes, 13 percent had positive internal iliac nodes only,2 which would have been missed by limited dissection. By comparing the results of magnetic resonance imaging (MRI) with lymphotropic superparamagnetic nanoparticles with the results of limited dissection, Harisinghani et al. may have overestimated the sensitivity of this MRI technique in detecting histologically positive nodes. Similarly, the use of only the length of the short axis, rather than the radiologist's clinical judgment, to define positive nodes on noncontrast MRI may not reflect the sensitivity and specificity obtained in clinical practice. In a study evaluating the effectiveness of contrast MRI in detecting nodal metastases, the use of a population of patients undergoing an extensive nodal dissection would have been optimal.
2 ReferencesRobert J. Yaes, Sc.D., M.D.
15 Quantum Pl., Gaithersburg, MD 20877
yaesr@cder.fda. gov 1
Harisinghani MG ,Barentsz J ,Hahn PF , et al. Noninvasive detection of clinically occult lymph-node metastases in prostate cancer. N Engl J Med 2003;348:2491-2499
Full Text | Web of Science | Medline2
Bader P ,Burkhard FC ,Markwalder R ,Studer UE . Is a limited node dissection an adequate staging procedure for prostate cancer? J Urol 2002;168:514-518
CrossRef | Web of Science | Medline
To the Editor:
Harisinghani et al. report that in five patients, the presence or absence of nodal metastases in specific lymph nodes was ascertained by computed tomography (CT)–guided biopsy, and no surgery was performed. What was the rationale for the use of this approach in these five patients? It is possible that CT-guided biopsy introduced a bias into the study, since only the larger, more suspicious nodes would have been sampled with the use of CT. Any cancer in smaller nodes would not have been sampled and might have been missed. Therefore, the authors should provide data about the size of the lymph nodes and the biopsy results in these five patients and what kind of treatment was used.
Ferdinand Frauscher, M.D.
University Hospital Innsbruck, 6020 Innsbruck, Austria
ferdinand.frauscher@uibk. ac. at Ethan J. Halpern, M.D.
Thomas Jefferson University, Philadelphia, PA 19107Andrea Klauser, M.D.
University Hospital Innsbruck, 6020 Innsbruck, AustriaAuthor/Editor Response
All patients enrolled in our study were eligible for surgical prostatectomy — that is, they did not have clinical or CT evidence of nodal metastases. In 5 of the 80 patients (6 percent of the study group), lymphotropic nanoparticle–enhanced MRI suggested the presence of solitary metastases in selected nonenlarged lymph nodes (mean nodal diameter, 7.8 mm). These specific lymph nodes were then sampled by CT-guided biopsy, revealing metastases in all five patients. Because of the positive pathological findings, these patients did not undergo additional surgery. These results indicate that patients with solitary lymph-node metastases may be spared surgical exploration.
The goal of our study was not to determine the correct extent of surgical lymph-node dissection, but rather to determine the accuracy of lymphotropic nanoparticle–enhanced MRI in the staging of prostate cancer. We agree with the assessment by Dr. Yaes that more extensive dissection will yield more lymph nodes and will therefore reveal additional metastases (surgery has a higher sensitivity).1 Because more extensive dissection is also associated with higher morbidity,2 there is some debate over the appropriate extent of dissection of pelvic lymph nodes.2 Furthermore, half of patients with newly diagnosed prostate cancer consider nonsurgical approaches to treatment (i.e., radiation therapy, observation, or cryotherapy), and they are deprived of an accurate evaluation of their lymph-node status.
Our study shows that the criteria used in traditional image analysis (size, shape, and clustering) have low accuracy,3 and that nanoparticle-enhanced MRI has exquisitely high sensitivity. Because of the high negative predictive value of the technique we described (100 percent) in a patient-by-patient analysis and 97.8 percent in a node-by-node analysis), one can argue that in patients with negative MRI studies, no dissection should be performed at all.4 In patients with positive MRI findings in small nodes (or nodes in the internal iliac region), either a more extensive dissection or a CT-guided biopsy may be performed. Because most modern MRI systems allow extended pelvic coverage at high spatial resolution, an MRI staging procedure also surveys lymph nodes far beyond the fields of extended surgical resection; this may be of additional benefit to patients and may improve their chance for a cure.
4 ReferencesMukesh Harisinghani, M.D.
Harvard Medical School, Boston, MA 02115Jelle Barentsz, M.D., Ph.D.
University Medical Center, Nijmegen 6500 HB, the NetherlandsRalph Weissleder, M.D., Ph.D.
Harvard Medical School, Boston, MA 02115
weissleder@helix.mgh. harvard. edu 1
Bader P ,Burkhard FC ,Markwalder R ,Studer UE . Is a limited node dissection an adequate staging procedure for prostate cancer? J Urol 2002;168:514-518
CrossRef | Web of Science | Medline2
Stone NN ,Stock RG ,Unger P . Laparoscopic pelvic lymph node dissection for prostate cancer: comparison of extended and modified techniques. J Urol 1997;158:1891-1894
CrossRef | Web of Science | Medline3
Jager GJ ,Barentsz JO ,Oosterhof GO ,Witjes JA ,Ruijs SJ . Pelvic adenopathy in prostatic and urinary bladder carcinoma: MR imaging with a three-dimensional TI-weighted magnetization-prepared-rapid gradient-echo sequence. AJR Am J Roentgenol 1996;167:1503-1507
Web of Science | Medline4
Meng MV ,Carroll PR . When is pelvic lymph node dissection necessary before radical prostatectomy? A decision analysis. J Urol 2000;164:1235-1240
CrossRef | Web of Science | Medline
