Annals of Oncology 10: 865-866, 1999. Letter to the editor Does pain at tumor site during vinorelbine infusion affect treatment of recurrent head and neck cancer patients? Vinorelbine (VNR) is a semisynthetic vinca alkaloid with selective activity against microtubules, which has proven to be active in several solid tumor types [1], including head and neck cancer [2], with a manageable toxicity. VNR is associated with a low incidence of severe adverse events including dose-limiting neutropenia, peripheral neuro- pathy and injection-site reactions. Beginning in December 1996, 17 patients with head and neck cancer metastatic or recurrent after radiotherapy (15 of 17) and/or chemotherapy (17 of 17) received salvage monoche- motherapy with VNR. A total of 86 courses were administered at the dose of 25 mg/m 2 over 20 minutes on days 1 and 8 every three weeks. In order to prevent emesis, ondansetron 8 mg i.v and/or metoclopramide 10 mg i.v. were delivered before each VNR infusion. Most patients had locally advanced disease with bulky lesions (latero-cervical swelling with skin infiltration) (10 cases), seven patients had metastatic disease (two of them with bulky lesions as well) (Table 1). None of the patients had previous medical histories of neurological or rheumatological disease, or signs of neurotoxicity. Six patients (35%), all with locally advanced disease and pretreated with radiotherapy and two or more chemotherapy regimens, experienced acute, very intense, sharp and distress- ing pain at the tumor site (head, neck or shoulder areas) within a few minutes after the first VNR administration. Concurrently a related vagal syndrome (lavish sweat, pallor, and shivering) was observed in four of six patients. In all cases the pain required discontinuation of the VNB infusion and administra- tion of an empirical treatment with steroids, ketorolac and/or tramadol. In every case the pain was reversible within half an hour after medication. One patient was admitted to the hospital because of a concomitant retrosternal pain that simu- lated a myocardial infarction. Table 1. Patient characteristics. Characteristics Number Sympto- of matic patients patients Total no. Locally advanced (all with bulky disease) Metastatic (2 also with bulky disease) Previous RT Previous CT One line regimen Two or more regimens Primary tumor site Oral cavity Larynx Nasopharynx Salivary glands 17 10 7 15 17 4 13 10 4 2 1 6 6 - 6 - 6 4 2 - - Five of these patients were further rechallenged with a second administration of VNB with analgesic premedication: three patients continued treatment without side effects, two experienced the same symptom that had led to treatment discontinuation. These two patients had greater loco-regional extension of the disease, together with skin infiltration and fistulae. One patient refused chemotherapy after the first ad- ministration. About 17% of the patients could not be treated withVNRatall. Among the few published experiences, Gebbia et al. [3] reported this symptom in a larger cohort of patients (12 of 150, 8%) with various types of cancer in whom VNR was always administered in combination with high doses of cis- platin, so the possibility of an interaction between the two drugs could not be excluded. Colleoni et al. [4] described an acute tumor pain which occurred after the end of VNR infusion in 10 of 135 patients (7%) with miscellaneous types of cancer (but none of them head and neck carcinoma), suggest- ing that there may be no preferential site for this syndrome. In the report by Kornek [5] a high rate of head and neck cancer patients 9 of 26 (34%) experienced the same symptom after 30 mg/m 2 VNB i.v. infusion and before the administration of a second cytotoxic agent. It is important to underline that in our series this event was peculiar to patients with gross disease, highly vascularized, with tissue damage, oedema, and probable necrosis. These findings suggest a possible relationship between the administration of this powerful vescicant drug and the vascular tumor bed, especially at bulky, ulcered sites such as a 'local extravasation' at tumor site. All patients were asymptomatic before treat- ment, none had received any other potentially neurotoxic drug, and VNR was given as monochemotherapy, so we are con- fident that the acute pain experienced was related exclusively to the administration of this drug. We observed no similar events in patients with other neoplasms. This unexpectedly high incidence of acute severe pain at the tumor site during VNR infusion, even at the minimum recommended dosages, should be borne in mind when dealing with patients with bulky recurrent head and neck cancer, in which this drug is moderately active, since it can adversely affect treatment delivery. Identification of new salvage agents in this subset of patients probably needs to be considered. S. De Marco, A. Fabi, A. Ceribelli, P. Carlini, C. F. Pollera & F. Cognetti 1st Division of Medical Oncology, Regina Elena Cancer Institute, Rome, Italy References 1. Cros S, Wright M, Morimoto C et al. Experimental antitumor activity of navelbine (5'-anydrovinblastine, vinorelbine). Sem Oncol 1989; 16 (Suppl 4): 15-20. 2. Degardin M, Oliveira J, Geoflrois Let al. An EORTC-ECSG phase II study of vinorelbine in patients with recurrent and/or metastatic squamous cell carcinoma of the head and neck. Ann Oncol 1998; 9: 1103-1107. 3. Gebbia V, Testa A, Valenza R et al. Acute pain syndrome at tumor site in neoplastic patients treated with vinorelbine: Report of unusual toxicity. Eur J Cancer 1994; 30A: 6,889. by guest on October 17, 2011 annonc.oxfordjournals.org Downloaded from