| ||||
| ||||
 Medscape |
 Home Site Map Marketplace My Medscape CME Center Feedback Help Desk |
| |
American Society of Clinical Oncology 35th Annual
Meeting
Day 2 - May 16,
1999
Marshall Posner, MD
Chemoprevention, the use of drugs to moderate and avert the development of cancer in high-risk patients, has been studied in head and neck cancer treatment for a decade and a half. Its effectiveness has yet to be determined, however.
Chemoprevention studies are long-term investigations requiring large pools of patients so that researchers may collect a sufficient number of events upon which to base their analyses. If the predicted number of events were too low, it would be unclear whether the trial had enough power to demonstrate the effectiveness of chemoprevention.
The first data coming from this randomized, placebo-controlled chemoprevention trial of 13-cis retinoic acid (CRA) in curatively treated, early-stage squamous cell cancer of the head and neck (SCCHN) were presented by Dr. Khuri and associates.[1] This trial was designed to determine whether chemoprevention with a high dose (30 mg/day) of CRA will prevent the development of second primary cancers of the upper aerodigestive tract (UADT), which includes the head and neck, esophagus, and lungs, in patients with potentially cured SCCHN.
The trial was initiated in 1991, recently achieved the planned accrual of 1120 patients, and will be closed on 6/30/99 for further accrual. Final results and unblinding will take place shortly thereafter.
Data on the development of second primaries and recurrences, updated for this meeting, were presented. Of importance, the second primary rate in this population of patients would be expected to be about 5% per year. To date, there have been 240 end points, defined as relapse or second primaries. The second primaries are defined as occurring more than 3 years after the initial primary if at the same site or more than 3 cm from the original primary. There have been 140 second primaries, of which 94 are in the UADT. Additional second primaries occurred in the bladder and the prostate.
The relationship between smoking behavior and UADT second primaries was studied in the evaluable patients. Evaluable patients were divided into 3 groups: never smokers (147 patients), former smokers having quit more than 1 year ago (537), and active smokers (287). In addition, there were 224 patients who were defined as not using alcohol or cigarettes. Cotinine levels confirmed nonsmoking status.
The average rate of second primaries was 4.9% per year in smokers compared with 3.3% in never smokers. This difference was highly significant. In addition, there was evidence that the rate was highly site-specific, being greater outside the larynx. The second primary rate was greater for patients presenting with stage II tumors, with smokers having a 2-fold greater incidence than nonsmokers. Second primaries were seen more frequently in stage II than in stage I cancers. Given recent data suggesting that some of these might be molecularly diagnosed as recurrences, an effort is being undertaken to obtain tissue to study this possibility.
While many might feel that a higher occurrence of second primaries in smokers is to be expected, it is important to note the high rate of second primaries in never smokers. At a rate of 3.3% per year, second UADT cancers might be expected to develop in almost 15% of nonsmokers in 5 years.
Also alarming is evidence that the recurrence rate is significantly higher in smokers than in nonsmokers. It is known that SCCHN patients who continue to smoke do poorly with therapy, but this large-scale documentation of increased recurrences in active smokers has biological implications. Smoking, and particularly nicotine, may have affected the tumor-cell genetics or the tumor environment, such as tumor vasculature, as well as other parameters.
This is a seminal study of chemoprevention; it is hoped that at next year's ASCO meeting we will be able to hear a preliminary analysis of the unblinded data. Given the intense scrutiny of the patients on this trial, we can expect new data and continuing surprises from this study.
Dr. Gillison and colleagues[2] presented data on Sunday documenting that HPV is associated with a subset of SCCHN. They analyzed 260 snap-frozen biopsy specimens obtained between 1988 and 1999 for HPV DNA. HPV was identified in 25% of samples. The predominant HPV subtype was HPV 16. HPV presence was confirmed by multiple methods. Southern blots identified HPV DNA in tumor lysates, and in situ hybridization demonstrated HPV restricted to tumor cell and not stroma.
Because there are subtypes of HPV 16 that can be identified by sequence analysis of the E6 gene, Gillison and colleagues were able to show that tumor infection was related to a single variant in each tumor. This latter finding, while not demonstrating clonality of HPV in tumors, is highly suggestive that HPV is an early event in carcinogenesis.
Of importance, HPV-positive tumors were more frequent in the oropharynx, involving predominantly the base of the tongue and tonsil, areas with a concentration of lymphoid tissues. The tumor pathology appeared to be "basaloid" squamous carcinoma, which is similar to the basaloid histology seen in cervical carcinoma. Of interest, p53 appeared to be wild type in the HPV-associated tumors, which is also consistent with the mode of p53 inactivation associated with HPV infection. In data presented in the abstract session, it was suggested that HPV might have a positive influence on survival that was independent of other factors.
Data from this and several other studies strongly suggest that HPV is responsible for a clinically significant proportion of SCCHN. HPV represents a possible therapeutic target in these patients. In addition, in those patients in whom the cancer may result from HPV, there may be a different prognosis, rate of second primaries, and response to therapy. HPV may join Epstein-Barr virus as a DNA virus known to be responsible for head and neck cancer.
Data were reported on response of injected lesions. There was a 54% response rate in all patients and a 63% response rate in evaluable patients. Also, the complete response (CR) rate was 27%. The response rate is higher than that expected from PF therapy alone. Also impressive was a local control rate of 83% in responding sites at 6 months. This prolonged time to progression would be unexpected from PF alone, in which a median time to progression in responders is less than 5 months. As further evidence of a viral-drug interaction, in several patients the injected tumors responded to the combination while other, uninjected sites progressed during the cycle (Table I).
Response No response Total Injected tumors 7 2 7/9 (78%) Uninjected tumors 3 6 3/9 (33%)
Biological evaluations showed continued viral replication in the tumors as late as 15 days after injection + chemotherapy, suggesting that there is continued viral-mediated tumor cell lysis taking place in the tumor. Viral replication was not observed in the adjacent normal tissues and stroma.
As a consequence of these results, a randomized Phase III evaluation of PF, with and without Onyx-015, is planned in SCCHN. These results are important because they are a necessary first step in developing this Phase III trial, which is the only way to demonstrate the effectiveness of combination agents. Onyx-015 is the first adenoviral product to enter Phase III testing in SCCHN.
All material on this website is
protected by copyright. Copyright ©
1994-1999 by Medscape Inc. All rights reserved. This website also contains
material copyrighted by 3rd parties. CME means Continuing Medical Education
credit is available. Medscape requires 3.x browsers or better from Netscape or Microsoft.
|
