MSLT-II, an international, multicenter, randomized, phase 3 trial to evaluate the usefulness of completion lymph-node dissection in patients with melanoma and sentinel-node metastases, consisted of a screening phase in which patients were enrolled before sentinel-node biopsy and a randomization phase in which completion lymph-node dissection was compared with observation and nodal ultrasonography (Fig. 1 ). The trial was conducted at 63 centers.
MSLT-II was designed by the MSLT-II executive committee with input from the pathology and ultrasonography oversight committees (see theSupplementary Appendix , available with the full text of this article at NEJM.org). Data were collected prospectively on paper and later on Web-based case-report forms. The authors vouch for the accuracy and completeness of the data and analyses reported and for the fidelity of the trial to the protocol, available at NEJM.org.
Nodal metastasis was determined by means of standard pathological assessment (including immunohistochemical tests performed according to institutional protocols) or by means of a previously described quantitative reverse-transcriptase–polymerase-chain-reaction (RT-PCR) assay during the screening phase.15 (link) Patients had to have undergone randomization and completion lymph-node dissection within 140 days after diagnostic biopsy.
The randomization phase involved enrollment of patients who had undergone screening and had pathologically or molecularly positive sentinel-node metastases and patients who had not undergone screening in whom sentinel-node metastases were detected by means of pathological assessment. In the randomization phase, fewer patients with RT-PCR–positive findings than anticipated were enrolled. In 2012, the data and safety monitoring board determined that such patients should no longer undergo randomization, since attainment of sufficient power to evaluate a therapeutic effect in that group was not feasible. The data and safety monitoring board recommended continued follow-up of these patients to assess outcomes.
At the third interim analysis, the data and safety monitoring board determined that detection of a significant survival difference between the trial groups was unlikely and recommended that the current primary end-point data be released. Intention-to-treat and per-protocol analyses of the outcome variables showed similar results. Results of per-protocol analyses are reported in this article, since they are likely to be the most clinically pertinent. The intention-to-treat data for the primary end point (melanoma-specific survival) are provided inFigure S1 in the Supplementary Appendix .
MSLT-II was designed by the MSLT-II executive committee with input from the pathology and ultrasonography oversight committees (see the
Nodal metastasis was determined by means of standard pathological assessment (including immunohistochemical tests performed according to institutional protocols) or by means of a previously described quantitative reverse-transcriptase–polymerase-chain-reaction (RT-PCR) assay during the screening phase.15 (link) Patients had to have undergone randomization and completion lymph-node dissection within 140 days after diagnostic biopsy.
The randomization phase involved enrollment of patients who had undergone screening and had pathologically or molecularly positive sentinel-node metastases and patients who had not undergone screening in whom sentinel-node metastases were detected by means of pathological assessment. In the randomization phase, fewer patients with RT-PCR–positive findings than anticipated were enrolled. In 2012, the data and safety monitoring board determined that such patients should no longer undergo randomization, since attainment of sufficient power to evaluate a therapeutic effect in that group was not feasible. The data and safety monitoring board recommended continued follow-up of these patients to assess outcomes.
At the third interim analysis, the data and safety monitoring board determined that detection of a significant survival difference between the trial groups was unlikely and recommended that the current primary end-point data be released. Intention-to-treat and per-protocol analyses of the outcome variables showed similar results. Results of per-protocol analyses are reported in this article, since they are likely to be the most clinically pertinent. The intention-to-treat data for the primary end point (melanoma-specific survival) are provided in
