Molecular Markers Are Used as Therapeutic Targets in Head & Neck Cancer: Three Studies

DENVER--Novel agents have been designed to exploit molecular markersas therapeutic targets in head and neck cancer. Two studies presented atthe American Society of Clinical Oncology meeting involved agents targetingthe p53 gene, either by eradication of p53-altered tumor cells or restorationof normal gene function. The third study involved a monoclonal antibodytargeted at tumor cells with an abnormality of the EGF receptor.

Daniel Von Hoff, MD, of the Cancer Therapy and Research Center and Universityof Texas Health Science Center, San Antonio, reported on ONYX-015, an attenuatedchimeric group C adenovirus construct with a deletion in E1B. The virusefficiently replicates in and lyses p53-deficient tumor cells but not p53-normalcells; p53 is altered in 50% to 70% of head and neck cancer patients.

The phase I study enrolled 27 patients with recurrent squamous cellhead and neck cancer and an abnormal p53 status by immunohistochemistry.All patients had failed standard surgery and radiotherapy, with or withoutchemotherapy.

The patients were given a single intra-lesional injection (10⁷or 10¹⁰ pfu) every four weeks. After safely reaching the 10¹⁰level, the researchers switched to a schedule of 10⁹ and 10¹⁰pfu daily times five.

The therapy was well tolerated with no grade 3 toxicity except for onecase of pain on injection. Nineteen single-dose patients were evaluablefor antitumor effects, but it was too early to evaluate the eight receivingdaily times five dosing.

Antitumor effects were measured in relation to the injected portionof the tumor, not the entire tumor burden. Using this criterion, "therewas definite evidence of an antitumor effect," Dr. Von Hoff said.

In one patient with a primary tongue lesion, 50% necrosis of the tumorwas observed with three cycles of treatment. "The patient had decreasedtrismus, improved jaw mobility, and improved swallowing with each successiveinjection," he said. Another patient with a left temporal primaryhad 75% tumor necrosis with two treatment cycles. In a third patient, witha tongue primary lesion, 80% of the injected area sloughed off eight daysafter treatment.

Dr. Von Hoff said that the researchers plan to continue to use the dailytimes five regimen "to try to address the previously uninjected portionsof the tumors." When the desired dose is determined, phase II efficacytrials will begin.

Restoring Normal p53 Function

Researchers from M.D. Anderson have attempted to restore p53 functionin patients with advanced recurrent squamous cell carcinoma of the headand neck by intratumoral injections of an adenoviral-mediated p53 gene(ADp53).

In this phase I study, previous chemotherapy for recurrent disease wasallowed. The recurrent tumor was evaluated for the presence of p53 mutationsby direct DNA sequencing, but mutations were not required for study entry.

All patients received the agent three times weekly for two consecutiveweeks. Then patients with resectable disease underwent surgery, with directinjection of ADp53 into the surgical bed. Three days later, an eighth dosewas given via retrograde catheter. Patients with nonresec-table diseasehad tumor biopsy in place of resection. The ADp53 dose was escalated from10⁶ pfu to 10¹¹ pfu.

Principal investigator Gary L. Clayman, MD, reported on 30 patients,70% with p53 abnormalities. The treatment was well tolerated, he said,with pain on injection the only grade 3 event.

ADp53 produced objective tumor regression, Dr. Clayman said, particularlyat the higher doses. Among the nonresec-table patients, there were no objectiveresponses at lower doses and two partial responses at 10¹⁰ and10¹¹ pfu.

One of the resected patients had a complete histologic response at 10⁷pfu. "No viable tumor was found in a completely resected neck specimen,"he said. Another resected patient treated at 10¹¹ pfu had anapproximately 30% reduction. Dr. Clayman noted that phase II studies arescheduled to begin later this year.

Growth factor can be stimulated by epidermal growth factor (EGF) ortransforming growth factor-alpha (TGF-alpha). EGF receptor is overexpressedin one third of solid tumors and is thought to play a pivotal role in bothinitiation and propagation of tumors, said Harlan W. Waksal, MD, chiefoperating officer, ImClone Systems Inc., New York.

Theoretically, a drug that binds to tyrosine kinase receptor would blockaccess of growth factor to the receptor and ultimately block cell proliferation.

Dr. Waksal reported on a phase Ib/IIa study of a chimerized monoclonalantibody that binds to the EGF receptor with an affinity two- to fivefoldhigher than that of EGF or TGF-alpha. The antibody, known as C225, wasdeveloped by John Mendelsohn, MD, of M.D. Anderson.

In preclinical studies, Dr. Waksal said, C225 was shown to synergizewith chemotherapy to inhibit tumor cell division. Thus, in this study ofadvanced head and neck and non-small-cell lung cancer patients, C225 wascombined with cisplatin.

The weekly IV C225 dose was escalated from 5 to 400 mg/m². Althoughpatients could not have received any therapy within one month prior tostudy, all but two had had some prior treatment.

"This was a very well tolerated drug," Dr. Waksal said. Therewere no dose-limiting toxicities and only one grade 3 toxicity--an anaphylactoidreaction.

At a C255 dose of 200 mg/m², Dr. Waksal said, "we hit whatwe are calling saturation of clearance." At this dose, he said, thehalf-life was 7 days and there was dramatic serum accumulation. The pharmacokineticsremained the same at 400 mg/m².

All three patients who received the 100 mg/m² dose had stable diseasefor 12 weeks. At 200 mg/m², two had stable disease and one had a partialresponse, although this was unconfirmed because the patient did not have30-day follow-up. At 400 mg/m², one of three patients had a partialresponse, which was confirmed at 30-day follow-up as a 57% reduction.

Dr. Waksal noted that both of the patients who responded had receivedprior radiation therapy but no chemotherapy.

He noted that ImClone is initiating other phase Ib/IIa studies to finetune the dose level at around 200 mg/m².

Limitations of the Studies

Everett Vokes, MD, the discussant for the papers, said that all threestudies had shown "proof of principle." However, he said, thetwo agents targeting p53 (ONYX-015 and ADp53) may have limited usefulnessfor two reasons: Because they are local therapies and "reasonable"local therapies already exist for head and neck cancer, and because theirefficacy may be limited to p53-altered tumor cells.

He suggested that these agents might be useful in combination with otherestablished modalities. "This is important in view of the contributionof p53 to the determination of resistance to chemotherapy or radiotherapy,"he said.

The EGF receptor blocker C255, studied in combination with cisplatin,showed fairly modest activity, he said, and the patients who respondedhad not received prior chemotherapy. "This kind of activity can certainlybe explained by cisplatin alone," he noted.