High-Dose Interferon to Treat Melanoma Offers No Benefit

CHICAGO — The Sunbelt melanoma study did not meet its primary end point of showing a benefit of high-dose interferon in melanoma patients with a single positive sentinel lymph node, reported the principal investigator at the annual meeting of the American Society of Clinical Oncology.

In neither of two protocols comparing high-dose interferon to observation did the treatment arms show a significant benefit in either 5-year disease-free survival or overall survival among patients with a single positive node confirmed by histology or molecular studies.

Dr. Kelly McMasters, chief of the division of surgical oncology at the University of Louisville (Ky.), presented the data on behalf of his colleagues in the Sunbelt Melanoma Trial Group. The investigator-initiated trial, which involved 79 centers in the United States and Canada, registered 3,619 patients from the ages of 18 to 70 years with cutaneous melanomas with a thickness of at least 1 mm and no clinical evidence of regional nodal or distant metastases. Of these 1,781 were reported in the intention-to-treat analyses.

The patients were stratified according to Breslow thickness: 1.0-2.0 mm, 2-4 mm, and greater than 4 mm, and the presence or absence of ulceration.

The trial had two protocol arms and an unusually complex design. All patients underwent sentinel node biopsy, and those with a node that was confirmed positive on histology were eligible for protocol A, which involved completion lymph node dissection.

Patients with only one microscopically positive node were then randomized to either an observation arm (arm 1, with 112 patients) or to a treatment arm with high-dose interferon alfa-2b (Intron A) for 12 months (arm 2, 106 patients).

Patients with more than one positive lymph node or extracapsular extension were assigned to receive high-dose interferon for 12 months and were followed (arm 3, with 99 patients).

In protocol B, those patients who had histologically negative results after sentinel node biopsy subsequently had their sample tested with reverse transcriptase-polymerase chain reaction (RT-PCR) for molecular staging to detect the presence of occult melanoma cells. Those who were RT-PCR-positive were then randomized to either observation (arm 4, with 180 patients), completion lymph node dissection (arm 5, with 192 patients), or completion lymph node dissection plus interferon (arm 6, with 184 patients). Patients who were RT-PCR-negative were assigned to observation alone (arm 7, with 908 patients).

Dr. McMasters presented intention-to-treat analyses for each protocol. In protocol A, for patients with a single histologically positive sentinel node, the 5-year disease-free survival was 70.2% for arm 1 (the observation group) and 73.2% for arm 2 (the interferon group) (log-rank P = .4589). Five-year overall survival in these patients was 75.4% among patients on observation vs. 72.9% for those on interferon (P = .9033).

There were, however, statistically significant decreases in both disease-free and overall survival for patients in protocol A with more than one positive lymph node (arm 3), compared with arms 1 and 2. The 5-year disease-free survival for patients in arm 3 was 44.5% (P less than .0001 vs. arms 1 and 2), and overall survival was 52.9% (P = .0004).

In the disease-free survival analysis in protocol B, there were similarly no significant differences among patients randomized to observation, completion lymph node dissection, or completion lymph node dissection with interferon, with rates of 83.9%, 85.2%, and 83.7%, respectively.

For the same groups in the overall survival analysis, there were no significant differences, with rates of 85.5% for the observation arm, 85.3% for the node dissection alone arm, and 86.8% for the node dissection plus IFN arm. Similarly, there were no significant differences in either disease-free survival in protocol B between patients who were node-negative or node-positive on RT-PCR, or in overall survival, which indicated that PCR testing for melanoma cells was not prognostically significant in this study, Dr. McMasters remarked.

A comparison of all the patient treatment arms in the study showed that patients who were histologically node-negative (those in arms 4, 5, 6, and 7 combined) had a better overall survival rate than those with a single positive sentinel node (patients in arms 1 and 2 combined), who in turn did better than those who had more than one positive node or extracapsular extension (those in arm 3) (P less than .0001).

In protocol A, "our results did not support the use of high-dose interferon for patients with a single microscopically positive sentinel node," Dr. McMasters said.

In protocol B, the results do not support the use of completion lymph node dissection or interferon for node-negative patients, and indicate that RT-PCR staging of sentinel lymph nodes was not predictive of worse outcome.

Dr. McMasters noted that the study was underpowered to detect small differences in disease-free survival and overall survival, "but we also did not observe significant trends, and even large sample size will not make differences appear if they don't exist."

The study was supported by a grant from Schering Oncology Biotech. Dr. McMasters and several other investigators have been on the speakers bureau.

ELSEVIER GLOBAL MEDICAL NEWS

CHICAGO — The Sunbelt melanoma study did not meet its primary end point of showing a benefit of high-dose interferon in melanoma patients with a single positive sentinel lymph node, reported the principal investigator at the annual meeting of the American Society of Clinical Oncology.

In neither of two protocols comparing high-dose interferon to observation did the treatment arms show a significant benefit in either 5-year disease-free survival or overall survival among patients with a single positive node confirmed by histology or molecular studies.

Dr. Kelly McMasters, chief of the division of surgical oncology at the University of Louisville (Ky.), presented the data on behalf of his colleagues in the Sunbelt Melanoma Trial Group. The investigator-initiated trial, which involved 79 centers in the United States and Canada, registered 3,619 patients from the ages of 18 to 70 years with cutaneous melanomas with a thickness of at least 1 mm and no clinical evidence of regional nodal or distant metastases. Of these 1,781 were reported in the intention-to-treat analyses.

The patients were stratified according to Breslow thickness: 1.0-2.0 mm, 2-4 mm, and greater than 4 mm, and the presence or absence of ulceration.

The trial had two protocol arms and an unusually complex design. All patients underwent sentinel node biopsy, and those with a node that was confirmed positive on histology were eligible for protocol A, which involved completion lymph node dissection.

Patients with only one microscopically positive node were then randomized to either an observation arm (arm 1, with 112 patients) or to a treatment arm with high-dose interferon alfa-2b (Intron A) for 12 months (arm 2, 106 patients).

Patients with more than one positive lymph node or extracapsular extension were assigned to receive high-dose interferon for 12 months and were followed (arm 3, with 99 patients).

In protocol B, those patients who had histologically negative results after sentinel node biopsy subsequently had their sample tested with reverse transcriptase-polymerase chain reaction (RT-PCR) for molecular staging to detect the presence of occult melanoma cells. Those who were RT-PCR-positive were then randomized to either observation (arm 4, with 180 patients), completion lymph node dissection (arm 5, with 192 patients), or completion lymph node dissection plus interferon (arm 6, with 184 patients). Patients who were RT-PCR-negative were assigned to observation alone (arm 7, with 908 patients).

Dr. McMasters presented intention-to-treat analyses for each protocol. In protocol A, for patients with a single histologically positive sentinel node, the 5-year disease-free survival was 70.2% for arm 1 (the observation group) and 73.2% for arm 2 (the interferon group) (log-rank P = .4589). Five-year overall survival in these patients was 75.4% among patients on observation vs. 72.9% for those on interferon (P = .9033).

There were, however, statistically significant decreases in both disease-free and overall survival for patients in protocol A with more than one positive lymph node (arm 3), compared with arms 1 and 2. The 5-year disease-free survival for patients in arm 3 was 44.5% (P less than .0001 vs. arms 1 and 2), and overall survival was 52.9% (P = .0004).

In the disease-free survival analysis in protocol B, there were similarly no significant differences among patients randomized to observation, completion lymph node dissection, or completion lymph node dissection with interferon, with rates of 83.9%, 85.2%, and 83.7%, respectively.

For the same groups in the overall survival analysis, there were no significant differences, with rates of 85.5% for the observation arm, 85.3% for the node dissection alone arm, and 86.8% for the node dissection plus IFN arm. Similarly, there were no significant differences in either disease-free survival in protocol B between patients who were node-negative or node-positive on RT-PCR, or in overall survival, which indicated that PCR testing for melanoma cells was not prognostically significant in this study, Dr. McMasters remarked.

A comparison of all the patient treatment arms in the study showed that patients who were histologically node-negative (those in arms 4, 5, 6, and 7 combined) had a better overall survival rate than those with a single positive sentinel node (patients in arms 1 and 2 combined), who in turn did better than those who had more than one positive node or extracapsular extension (those in arm 3) (P less than .0001).

In protocol A, "our results did not support the use of high-dose interferon for patients with a single microscopically positive sentinel node," Dr. McMasters said.

In protocol B, the results do not support the use of completion lymph node dissection or interferon for node-negative patients, and indicate that RT-PCR staging of sentinel lymph nodes was not predictive of worse outcome.

Dr. McMasters noted that the study was underpowered to detect small differences in disease-free survival and overall survival, "but we also did not observe significant trends, and even large sample size will not make differences appear if they don't exist."

The study was supported by a grant from Schering Oncology Biotech. Dr. McMasters and several other investigators have been on the speakers bureau.

ELSEVIER GLOBAL MEDICAL NEWS

CHICAGO — The Sunbelt melanoma study did not meet its primary end point of showing a benefit of high-dose interferon in melanoma patients with a single positive sentinel lymph node, reported the principal investigator at the annual meeting of the American Society of Clinical Oncology.

In neither of two protocols comparing high-dose interferon to observation did the treatment arms show a significant benefit in either 5-year disease-free survival or overall survival among patients with a single positive node confirmed by histology or molecular studies.

Dr. Kelly McMasters, chief of the division of surgical oncology at the University of Louisville (Ky.), presented the data on behalf of his colleagues in the Sunbelt Melanoma Trial Group. The investigator-initiated trial, which involved 79 centers in the United States and Canada, registered 3,619 patients from the ages of 18 to 70 years with cutaneous melanomas with a thickness of at least 1 mm and no clinical evidence of regional nodal or distant metastases. Of these 1,781 were reported in the intention-to-treat analyses.

The patients were stratified according to Breslow thickness: 1.0-2.0 mm, 2-4 mm, and greater than 4 mm, and the presence or absence of ulceration.

The trial had two protocol arms and an unusually complex design. All patients underwent sentinel node biopsy, and those with a node that was confirmed positive on histology were eligible for protocol A, which involved completion lymph node dissection.

Patients with only one microscopically positive node were then randomized to either an observation arm (arm 1, with 112 patients) or to a treatment arm with high-dose interferon alfa-2b (Intron A) for 12 months (arm 2, 106 patients).

Patients with more than one positive lymph node or extracapsular extension were assigned to receive high-dose interferon for 12 months and were followed (arm 3, with 99 patients).

In protocol B, those patients who had histologically negative results after sentinel node biopsy subsequently had their sample tested with reverse transcriptase-polymerase chain reaction (RT-PCR) for molecular staging to detect the presence of occult melanoma cells. Those who were RT-PCR-positive were then randomized to either observation (arm 4, with 180 patients), completion lymph node dissection (arm 5, with 192 patients), or completion lymph node dissection plus interferon (arm 6, with 184 patients). Patients who were RT-PCR-negative were assigned to observation alone (arm 7, with 908 patients).

Dr. McMasters presented intention-to-treat analyses for each protocol. In protocol A, for patients with a single histologically positive sentinel node, the 5-year disease-free survival was 70.2% for arm 1 (the observation group) and 73.2% for arm 2 (the interferon group) (log-rank P = .4589). Five-year overall survival in these patients was 75.4% among patients on observation vs. 72.9% for those on interferon (P = .9033).

There were, however, statistically significant decreases in both disease-free and overall survival for patients in protocol A with more than one positive lymph node (arm 3), compared with arms 1 and 2. The 5-year disease-free survival for patients in arm 3 was 44.5% (P less than .0001 vs. arms 1 and 2), and overall survival was 52.9% (P = .0004).

In the disease-free survival analysis in protocol B, there were similarly no significant differences among patients randomized to observation, completion lymph node dissection, or completion lymph node dissection with interferon, with rates of 83.9%, 85.2%, and 83.7%, respectively.

For the same groups in the overall survival analysis, there were no significant differences, with rates of 85.5% for the observation arm, 85.3% for the node dissection alone arm, and 86.8% for the node dissection plus IFN arm. Similarly, there were no significant differences in either disease-free survival in protocol B between patients who were node-negative or node-positive on RT-PCR, or in overall survival, which indicated that PCR testing for melanoma cells was not prognostically significant in this study, Dr. McMasters remarked.

A comparison of all the patient treatment arms in the study showed that patients who were histologically node-negative (those in arms 4, 5, 6, and 7 combined) had a better overall survival rate than those with a single positive sentinel node (patients in arms 1 and 2 combined), who in turn did better than those who had more than one positive node or extracapsular extension (those in arm 3) (P less than .0001).

In protocol A, "our results did not support the use of high-dose interferon for patients with a single microscopically positive sentinel node," Dr. McMasters said.

In protocol B, the results do not support the use of completion lymph node dissection or interferon for node-negative patients, and indicate that RT-PCR staging of sentinel lymph nodes was not predictive of worse outcome.

Dr. McMasters noted that the study was underpowered to detect small differences in disease-free survival and overall survival, "but we also did not observe significant trends, and even large sample size will not make differences appear if they don't exist."

The study was supported by a grant from Schering Oncology Biotech. Dr. McMasters and several other investigators have been on the speakers bureau.

ELSEVIER GLOBAL MEDICAL NEWS