Changing the pathophysiology of solid tumours: The potential of TNF and other vasoactive agents

Abstract

Solid tumour therapy with chemotherapeutic drugs is hampered by a number of factors resulting in poor results and failure of initially promising drugs. From the application of Tumour Necrosis Factor α in the melphalan-based Isolated Limb perfusion some lessons could be learned. Most importantly, combination of treatment approaches, certainly when multiple targets are involved, increases the effectiveness of the therapy. Clinical outcome may improve dramatically when the tumour pathophysiology is changed in such a way that co-administered chemotherapeutics are more active or are capable of reaching the tumour cells better. Here some of the methodologies and drug combinations which improve solid tumour therapy through acting on the tumour pathophysiology are discussed.

• Solid tumour
• tumour pathophysiology
• regional therapy
• perfusion
• TNF
• vasoactive drugs

Introduction

Tumour necrosis factor α (TNF), isolated 30 years ago, is a multi-functional cytokine playing a key role in apoptosis and cell survival as well as in inflammation and immunity. Although named after its anti-tumour properties TNF has been implicated in a wide spectrum of other diseases. The current use of TNF in cancer is regional treatment of locally advanced soft tissue sarcomas and large, bulky melanomas. In these patients TNF seems to act synergistically when combined with cytostatic drugs.

Interaction of TNF with TNF Receptor-1 (TNF-R1) activates a number of pathways leading to the diverse functions of TNF. The molecules engaged in TNF-R1 signalling have been elucidated quite well, but regulation of signalling is not fully understood. Besides these molecular insights, laboratory experiments in the past decade have shed light upon TNF action during tumour treatment. Next to induction of erythrocyte and lymphocyte extravasation, leading to haemorrhagic necrosis, TNF seems to target the tumour-associated vasculature (TAV) by inducing hyper-permeability and destruction of the vascular lining. This study provided evidence that this results in a selective accumulation of co-administered cytostatic drugs inside the tumour. It was hypothesized that these TNF-related observations may explain the success of TNF-based Isolated Limb Perfusions (ILP) in patients with irresectable soft tissue sarcomas (STS), bulky melanomas and other limb threatening tumours.

Small melanoma in-transit (IT)-metastases respond well to ILP with melphalan alone, but larger tumours and especially locally advanced STS are known to respond poorly to ILP with chemotherapeutics alone. This situation has changed completely by the utilization of TNF in the ILP setting. TNF-based ILP with melphalan appeared highly effective in multi-centre trial settings demonstrating response rates of above 70% and limb salvage rates of above 80%. This resulted in the approval of TNF in Europe for the ILP setting. In the laboratory other agents were identified, such as interleukin-2 and histamine, capable of enhancing the tumour response, which was accompanied by a similar effect on drug uptake in the tumour. Moreover, it was demonstrated that these agents may be more appropriate for isolated hepatic perfusions for the treatment of irresectable liver malignancies.

Next to loco-regional application of TNF, it was demonstrated that at low repeated doses this cytokine can be used systemically to enhance leakage of long circulating liposomes at the tumour site. This resulted in enhanced drug accumulation and synergistic anti-tumour response at TNF doses that are expected to be clinically applicable. Therefore, one urges that TNF be re-explored in the clinical setting in combination with liposomal drug formulations. This review provides an overview of the actual as well as the potential use of TNF and ‘like’ molecules in the treatment of cancer.

Delivery to solid tumours: Overcoming barriers

Solid tumour therapy is faced by a number of factors acting against it. While a drug may appear quite active in vitro, application in patients is often met by great disappointment, which may be due to side-effects which cannot be overcome, resistance of tumour cells to the drug in vivo or poor pharmacokinetics. Delivery of substantial amounts at the tumour cell site is of principal importance in cancer chemotherapy. A drug, when injected systemically into the blood stream or given orally, has to take a number of hurdles to reach the tumour cell [1]. Modifications can be made to improve efficacy of therapy. An important step forward is to define targets closer to the injected drug, i.e. the endothelial lining, which comprises the first cells seen by an agent after intravenous or intra-arterial administration. Secondly, vascular manipulation can be used to help drugs, which act on tumour cells, to better reach their target. Next to that, tumour response would greatly benefit from enhanced local drug levels, which can be accomplished by drug targeting (e.g. with the aid of liposomes) or by limiting dilution of the drug and at the same time increasing local drug level, which can be accomplished in a so-called isolated perfusion setting. Below examples of these modalities are discussed with emphasis on combining chemotherapeutic drugs with manipulation of the tumour vascular endothelial lining.

Improving delivery to solid tumours

One very straightforward method to increase drug delivery to the tumour cell is by enhancing the drug concentration around the tumour. It has been demonstrated that solid tumours confined to limb or liver can be treated with increased efficacy by a so-called isolated perfusion [2]. In such a setting the area in which the tumour resides is temporarily isolated from the body by clamping vessels and connecting these to a perfusion system, typically including a heart–lung type of device, which was originally pioneered for limbs by Creech et al. [3]. This methodology not only improves local concentration [4], but also diminishes systemic exposure allowing higher dosages to be used, especially when treatment is followed by a thorough washout of the perfused area. Typically melphalan is used in the ILP setting as other drugs appear less effective and often are associated with increased regional toxicity [5], [6]. Also doxorubicin, which is thought to be the most effective drug against sarcoma, is used by some groups in ILP, although it is believed to have higher local toxicity [7]. To further enhance efficacy of the ILP procedure application of a double perfusion, with a 3–4 week interval between perfusions, has been proposed. Although the double-perfusion schedule resulted in more complete remissions, no differences were observed in recurrence rates in the two groups [8]. Also combination of chemotherapy regimens at the same time in ILP did not render results encouraging enough in patients with irresectable tumours [9].

Secondly, combination treatment of solid tumours may boost response rates as well, especially when different targets are approached. For instance, tumour response and thus limb salvage is enhanced in the Isolated Limb Perfusion (ILP) setting by addition of the cytokine TNF to the chemotherapeutic drug melphalan [2], [10], [11]. It is hypothesized that TNF affects mainly the endothelial lining of the tumour-associated vasculature and by doing so inflicts disruption of the vascular bed in the tumour [12]. Importantly, TNF induces a six-fold augmented melphalan accumulation in the tumour, but not in healthy tissues such as skin and muscle [13]. The locally reached melphalan level consequently shifted from a non-effective concentration to one capable of killing tumour cells as determined in vitro. This double function of TNF is thought to be of primary importance for the observed improvement in response rate. As this has been shown in rat ILP models, with relatively small tumours compared to the human ILP setting, confirmation in patients is needed. The use of TNF is, therefore, an excellent example of a drug that primarily encounters a target closer to the drug, i.e. the endothelial cell, and secondly manipulates the tumour pathophysiology in such a way that the co-administered drug can work better.

Thirdly, local conditions can be adjusted in such a way to maximally favour tumour response. For instance, the degree of oxygenation, local pH-value or temperature in and around the tumour may affect response rate. Adjustment of oxygenation during ILP demonstrated that high pO₂ in the perfusate improved muscle pH. However, this may have an unwanted effect on efficacy as it has been shown that low pH increases the cytotoxicity of melphalan [14], [15]. An optimum has been observed in local temperature which may also have a direct effect on the tumour response [16].

Tumour vascular manipulation

As indicated above TNF is capable of affecting the tumour-associated vasculature in such a way that an increased leakage of the co-administered drug occurs. It is hypothesized that TNF may act on endothelial cells through functional down modulation of the integrin αvβ3 resulting in detachment and subsequent death of the endothelial cells [17]. It was found that TNF enhances melphalan and doxorubicin accumulation in tumour tissue during an ILP 3–6-fold in a time frame of 30 min. Although TNF exhibits important and dose limiting toxicity the use of low dose TNF in a systemic combination treatment setting was explored. Also at concentrations too low to have a direct cytotoxic effect on tumour or endothelial cells, an augmented drug accumulation of Doxil (liposomal formulated doxorubicin) was reached, resulting in dramatic improvement of tumour response [18], [19]. Strikingly, both at high dose (in ILP) and low dose (systemic) TNF did not seem to affect the endothelial lining of healthy tissues. Is seems that TNF induces an oedema-like response primarily in tumour tissue, which is accompanied by haemorrhagic necrosis when used at high concentrations. This observation initiated the exploration of other agents with comparable activity but with less intrinsic toxicity.

Application of vaso-active agents

As stated above local temperature greatly affects tumour response, which could be due to a dilation of the tumour vessels and thus a better delivery of the perfused drug. More so it was shown that hyperthermia may affect tumour and in particular endothelial cells directly [20].

Based on the observations with TNF, agents with comparable characteristics were examined concerning induction of vascular leakage. First, Histamine, an agent well known for its oedema inflicting capacity, was investigated. When histamine was used in the rat sarcoma ILP setting, massive tumour response was observed when combined with melphalan [21]. Moreover, Histamine is capable of disrupting the tumour-associated vasculture within a 30 min time frame, resulting in profound haemorrhagic necrosis. IL-2, a cytokine well known from immunotherapy and induction of oedema, was explored in the ILP setting [22]. While this agent seemed to have a more subtle effect on the endothelial lining, profound synergy was observed with melphalan in the ILP setting. Both Histamine and IL-2 augmented tumour drug accumulation significantly which was believed to greatly add to the improved tumour response. It was demonstrated that intra-tumoural melphalan level shifted from a hardly effective concentration to one well above the in vitro determined IC50 for sarcoma cells [22].

It is noteworthy to realize that tumours exhibit generally an increased interstitial fluid pressure (IFP) compared to for instance muscle [23], which can be lowered by the systemic administration of TNF [24]. However, no effect was observed of TNF on the IFP in tumours in the ILP setting (personal communication). While in the ILP setting reduction of the IFP may not work recent studies clearly show a potential use for agents with such effect [25–27]. However, this goes beyond the purpose of this review.

Conclusion

Solid tumour therapy greatly benefits from an enhanced delivery of chemotherapeutics at the tumour cell. By the use of loco-regional treatment methods, such as isolated perfusion, high drug levels can be reached in the tumour. Secondly, delivery can be improved by the use of vaso-active agents such as TNF, IL-2 and Histamine. It is important to note that the administration of a chemotherpeutic agent is essential to gain synergy. It is believed that combination of the surgical approach with chemotherapy and anti-tumour vascular therapy is a strong collaboration in the effective treatment of solid tumours.