A phase I study to study arsenic trioxide with radiation and hyperthermia in advanced head and neck cancer

Abstract

Purpose: Arsenic trioxide [ATO] is a pluripotent drug with potentials to have pro-oxidant, angiogenesis inhibitor, flow inhibitor and radiation sensitizer properties.

Methods: The present study is a Phase I trial to assess the safety of ATO in advanced or recurrent head and neck cancer treated with radiation and hyperthermia. Patients received ATO at 10, 20 and 30 mg per week a day prior to hyperthermia.

Results: It was assumed that vascular collapse would be complete by 24 h. Administration of ATO at 20 mg was safe with no toxicity due to ATO. No amplification of toxicities due to radiation or hyperthermia was evident. Patients without prior treatment showed better response. A total of 11 patients were included in this Phase I study.

Conclusions: Patients who received 30 mg of ATO weekly showed non-serious acute toxicities. No further escalation of dose was attempted.

• Arsenic trioxide (ATO)
• head and neck cancer
• hyperthermia
• radiation
• radiation sensitizer

Introduction

Patients with locally advanced head and neck squamous cell carcinoma (HNSCC) traditionally receive definitive radiotherapy. Over the past few decades, however, several approaches designed to increase the anti-tumour efficiency of radiotherapy of head and neck squamous cell carcinoma have been tested, including regimens combining radiotherapy with concomitant chemotherapy and radiation sensitizers [1]. Three meta-analyses of updated individual data MACH-NC shows a very moderate benefit with chemotherapy radiation in Collaborative group trials [2]. As an alternative to chemoradiation, Valdagni et al. [3] demonstrated effectiveness of hyperthermia added to radiation in head and neck cancer. However, further innovations are needed to improve response to thermoradiotherapy and increase survival of patients with advanced head and neck cancer. Verapamil and hyperosmolar dextrose have been tried to modulate vascular response to improve response following hyperthermia. Results have been equivocal. Arsenic trioxide is an old molecule showing new potential to be a thermal sensitizer as well as radiation sensitizer. Griffin et al. [4] have demonstrated arsenic trioxide (ATO) to induce selective tumour vasculature destruction via oxidative stress and increase thermosensitivity in a murine model [5]. ATO has recently received approval from FDA for the treatment of promyelocytic leukaemia [6].

There are ongoing Phase I–II clinical trials to test ATO in various solid tumours for its cytotoxic potential. ATO is a molecule with protean cellular effects [7–9]. It has been shown to induce apoptosis, growth inhibition, promotion of differentiation and inhibition of aniogenesis [10], [11]. ATO is shown to be a potent thermal enhancer due to its potential to induce vascular shutdown in solid tumours [12], [13].

The present study was designed to assess the toxicity of ATO, delivered weekly in conjunction with conventionally fractionated radiation therapy and loco-regional hyperthermia of patients with advanced or recurrent head and neck cancer.

Materials and methods

The current study is designed to evaluate the toxicity profile of ATO, delivered weekly in conjunction with radiation and hyperthermia. It was initiated in the year 2004 following a clearance from the hospital ethical committee. Patients with recurrent or advanced head and neck cancer were eligible for the study. All patients had biopsy-proven squamous cell carcinoma. Patients were recruited only if the Karnofsky's index was at least, or above, 70. All patients included for this study underwent base-line investigations, viz., haemogram, chest x-ray, serum creatinine and electrocardiogram. They were staged according to a TNMC (UICC) staging system. There was no age restriction for inclusion. Patients having failed previous surgery, chemotherapy and radiation were also eligible. The exclusion criteria for the study were history of recent myocardial infarction, ischemic heart disease or arrhythmia. Patients with neuropathies due to any co-morbid conditions like diabetes or prior chemotherapy were not eligible for inclusion. Patients with a known psychiatric illness were also not eligible. Patients with short and fat necks were also not eligible.

All patients were treated with parallel opposed compensated fields, or by 3-field technique, on Theratron 780C (Atomic Energy Canada Ltd.), a telecobalt machine. Patients were planned to receive a total dose of 60–70 Gy in 6–7 weeks with conventional fractionation. Hyperthermia for these patients was delivered on Saturdays, the day no radiation was given. ATO was administered 24 h prior to hyperthermia treatment. Hyperthermia was delivered using a radio-frequency heating machine (modified Thermatron RF8). Patients were positioned on the couch in supine position.

A pair of parallel electrodes of size 10 cm was placed across the neck. An active electrode was placed on the side of the disease. Patients underwent pre-cooling for 10–15 min at a temperature of 5°C. However, temperature of circulating water in the bolus was kept at 20°C during the treatment. Invasive thermometry with a single thermocouple was performed at least once during the treatment. A 19 guage angiocatheter was inserted in the metastatic node, taking due aseptic precautions. The thermocouple was inserted in the catheter till a resistance was felt. Insertion was done under vision. The insertion site varied from 2–3 cm from the surface. Increments in temperature were measured at an interval of 60 s and recorded. The treatment was deemed adequate if patients achieved a temperature of 40°C for at least 15 min. Arsenic trioxide was administered 1 day before the hyperthermia treatment on every Friday. Arsenic trioxide was delivered as an intravenous infusion over 3 h. Patients were pre-medicated with 8 mg of Ondansetron.

ATO is available as ‘Arsenox’ and is marketed by INTAS, Biotechnology Oncology, India. Incremental doses of 10, 20 and 30 mg were administrated. Dose was escalated after every four patients until prohibitive toxicity was noticed. Patients were monitored for treatment-related toxicity such as mucositis, thermal burns, gastrointestinal symptoms, blood pressure and heart rate; ECG was done only if clinical condition required it for assessment. Acute toxicities were assessed twice a week during the treatment. Mucosal toxicity was graded according to WHO guidelines into grade ‘0’ to grade IV and thermal burns were graded as grade I–III, depending on depth of burn, grade I being superficial and grade III being full-thickness involvement of skin.

Patients were assessed for response at the end of the treatment and periodically thereafter. Patients underwent clinical evaluation every 3 or 4 months. Appropriate imaging was requisitioned if necessary.

Results

A total of 11 patients were recruited for the study, of which eight patients were male and three female. Age varied from 41–71 years, with an average of 63 years. All the patients had histopathologically-proven squamous cell carcinoma. Four of the patients had anterior chemotherapy with cisplatinum and 5 fluorouracil. These patients had received chemotherapy at other hospitals and were then referred to the centre. Only one patient had undergone radical radiotherapy with weekly hyperthermia and weekly chemotherapy with 60 mg of Taxol before starting on re-induction with ATO radiation and hyperthermia. She had an aggressive cancer, which had recurred within 3 months. Table I shows the clinical profile, details of treatment and acute toxicities.

Table I.  Clinical profile of patients and treatment outcome.

Age (years)/sex	Diagnosis and stage	Prior treatment	Radiation dose (cGy)	ATO–cumulative (mg)	No. of HT session	Initial response	Toxicity (ATO)	Final outcome
53/M	Buccal mucosa, T4N0M0	Ant CT 1#	6600	10 × 6 = 60	6#	PR	None	Died, 2 months
41/F	Vocal cord recurrent, T2N3M0	CT + RT + HT	3600	10 × 4 = 40	4#	PR	None	Died, 3 months
65/M	Base tongue, T4N3M0	None	3800	10 × 2 = 20	2#	Incomplete	None	Died, 3 months
56/M	Pyriform fossa Rt. T2N3M0	Ant CT 3#	5800	10 × 2 = 20	2#	Incomplete	None	LF, 0 days
68/M	Pyriform fossa, T4N3M0	None	6600	20 × 7 = 140	7#	CR	None	Died, 1 month
79/M	Supraglottic Larynx, T4N0M0	None	6000	20 × 6 = 120	6#	CR	None	NED, 6 months
50/M	Unknown primary, TxN3M0	None	7000	20 × 5 = 100	5#	PR	None	NED, 7 months
64/M	Vocal cord, T3N2M1	Ant CT 4#	6400	20 × 5 = 100	5#	CR	None	NED, 8 months
68/M	Unknown primary, TxN3M0	Ant CT 1#	6000	30 × 2 = 60	6#	CR	Allergy	NED, 7 months
75/F	Lateral border tongue (Rt) T3N3M0	None	2400	30 × 2 = 60	2#	Incomplete	Sinus, Tachycardiasweating	Died, 15 days
67/M	Supraglottic Larynx, T4N3M0	None	7000	30 × 1 = 30	1#	PR	Restlessness,Tachycardia	LF, 1 month

CT, Chemotherapy; #, Fractions or frequency; NED, No evidence of disease; LF, Lost to follow-up.

Four patients in each group received 10 and 20 mg of ATO per week while three others received 30 mg per week. The cumulative dose of ATO received by patients varied from 30–120 mg. All the three patients who received 30 mg per week had ATO-related toxicity. One patient had urticaria and generalized rashes following ATO administration for two successive weeks, following which, both HT and ATO administration were stopped. The patient recovered on both occasions after the administration of cetrizine orally. An elderly woman of age 75 years developed uneasiness, palpitation and sweating. Electrocardiogram and blood pressure were within normal limits. Further treatment with ATO was interrupted on clinical grounds. The patient died 3 weeks later at home, of an unknown cause. A third patient also developed a similar symptom complex after the first injection of ATO. He, however, completed radiation without hyperthermia and ATO.

Discussion

There is increasing clinical evidence of hyperthermia and radiation being effective in solid tumours [14], [15]. The combination of hyperthermia and ionizing radiation has led to enhanced responses in different types of cancers such as cancer of the cervix and head and neck. Effective modulators of hyperthermia are scarce and include hyperosmloar dextrose and verapamil.

Arsenic trioxide is approved by the FDA for the treatment of acute pro-myelocytic leukaemia [7]. Other indications are being explored as in testicular tumours and glioma. It is commercially available as a sterile injectable solution without any preservative at a concentration of 1 mg ml⁻¹. The pharmacokinetics of the drug is poorly understood. The metabolism of ATO involves reduction of pentavalent arsenic to trivalent arsenic by arsenate reductase and methylation of trivalent arsenic to monotheye arsenic acid and monomethylarsonic acid to dimethylarsiuic acid by methyltroamsferaseses. The main site of methylation reaction appears to be liver. Excretion of ATO is predominantly via urine.

Arsenic trioxide is an old molecule with new indications that it may be an enhancer of hyperthermic effects. ATO has shown to be a potent anti-vascular agent that markedly potentiates the effect of hyperthermia on tumours. Griffin et al. [4] demonstrated a marked decrease in blood perfusion in SCK tumour of AIJ, mice and FS a_II tumours of C₃ H mice, 24 h after administering ATO intraperitoneally, at the rate of 8 mg kg⁻¹ of ATO. Griffin et al. have also demonstrated a dose-dependent decrease in perfusion at 2 h in Fsa_II and SCK tumours. Young et al. [5] demonstrated preferential vascular shutdown in fibrosarcoma growing in BALB/mice. A single administration of ATO (10 mg kg⁻¹) resulted in massive central necrosis and vascular shutdown at 2 and 6 h. In half of the mice, perfusion remained low even at 24 h.

Perfusion remained unchanged in skin, muscle and kidney. This preferential effect of ATO in tumour probably is due to structural faults in the vascular architecture of tumours [13].

Tumour blood flow is heterogeneous and follows a chaotic pattern of distribution with many constricted and sharp turns through vessels that often have gaps, as in endometrium, and lacks smooth muscle control [14]. ATO-induced stress is selective due to hypoxic micro-environment of the tumour as well as the increased expression of adhesive molecules. Hyperthermia is known to promote adhesion molecules and inflammatory cytokine production [16]. Hyperthermia with ATO can synergistically affect tumour vasculature to create greater vascular damage and flow inhibition [11–13]. The premise of this study is based on the above observation of ATO in vivo and in vitro.

One of the dilemmas of Phase I trial recruitment is inclusion of patients with extensive disease, prior treatment and, maybe, sometimes with co-morbid disease. Administration of any drug is fraught with much higher probabilities of toxicities in such patients. The deaths seen in this series are temporally far removed from the dates of ATO administration except for one patient who died while on treatment. Hence it is unlikely to be related to administration of ATO. None of the patients in this series manifested typical toxicities linked to high doses of arsenic trioxide or arsenic poisoning such as vomiting, abdominal cramps, delirium and sensory symptoms of pins and needles, myalgia and visual disorders. As little as 1–2.5 mg kg⁻¹ of arsenic trioxide administrated as a single dose can prove fatal. The highest administered dose in this Phase I trial is a fraction of the above suggested dose. It varied from 0.4–0.5 mg kg⁻¹ in the last group. The dose exploration was based on the review of various ongoing protocols using ATO. The present study recruited patients with recurrent tumours or loco-regionally advanced head and neck cancers. Patients tolerated the administration of ATO in graded doses of 10 and 20 mg administered intravenously over 2 h every week. No ATO-related toxicity was seen in these groups. Patients who received 30 mg of ATO weekly showed acute non-serious toxicities. The treatment was interrupted in one patient who developed tachycardia, profuse sweating and uneasiness. He had rapid pulse rate with stable blood pressure. Electrocardiogram following the second attack of sweating was within normal limits. He had received 30 mg of ATO. None of the patients showed ECG abnormalities including QT prolongation over 500 ms. No haematological or clinically evident neurological toxicities were seen in any patient. Table I shows acute toxicities due to ATO, radiation and hyperthermia.

This study shows that 20 mg of ATO can be safely administered as infusion every week with radiation and hyperthermia. The decision to sequence ATO followed by hyperthermia after 24 h was based on the study by Griffin et al., mentioned above. However, there is a gap in the complete understanding of vascular response following ATO. Hence, the next step in the study is to assess vascular dynamics after the administration of ATO in the clinical setting. This is a crucial step before starting a Phase II trail. A cumulative dose of 140 mg was tolerated very well. However, caution is warranted while administering dose at 0.4–0.5 mg kg⁻¹ of ATO weekly.

ATO has been tried as a single drug in the treatment of selected solid tumours. In a Phase II trial of hormone refractory prostate cancer, treatment with ATO (0.2 mg kg⁻¹ per day for two continuous 5 day periods on a 28-day cycle for one cycle followed by twice a week thereafter) resulted in a marked decrease in prostrate-specific antigen (PSA) in two of 15 assessable patients and slowed increase in PSA in another 12 patients. Acute toxicity in this study was acceptable [8]. Patients with metastatic renal cell carcinoma were treated with ATO at a dose of 0.3 mg kg⁻¹ per day for 5 consecutive days every 4 weeks. Response was not significant nor was any toxicity [9].

Conclusion

Arsenic trioxide was well tolerated in this Phase I study. Appropriate sequencing of HT and ATO administration with conventional fractionation needs further elucidation. The mechanistic basis of effectiveness of ATO is not clear. It could be due to its pleiotropic effects on cellular activities like cell cycle progression, DNA repair, ubiquination, tubulin ploymerization, nitric acid synthesis and oncogene expression or activation.