OUP user menu

Endothelin-1 and Metastatic Cancer Pain

Gudarz Davar MD
DOI: http://dx.doi.org/10.1046/j.1526-4637.2001.002001024.x 24-27 First published online: 1 March 2001


Pain in patients with metastatic cancer contributes to increased suffering in those already burdened by their advancing illness. The causes of this pain are unknown but likely to involve the action of tumor-associated mediators and their receptors. One such mediator, endothelin-1 (ET-1), can induce both pain-like behavior in animals and pain in humans that is endothelin-A (ETA) receptor-dependent, and that appears to be due to the selective excitation of pain fibers. More significantly, in clinical studies, antagonists of the ETA receptor have been shown to ameliorate pain in some patients with advanced metastatic prostate cancer. The identification of tumor-associated mediators such as ET-1 that might directly or indirectly cause pain in patients with metastatic disease should lead to improved, targeted analgesia for patients with advanced cancer.

  • endothelin
  • cancer pain
  • metastatic cancer

Pain associated with the spread of cancer from its primary site is often debilitating and difficult to treat, especially in patients with advanced disease. Among all cancer patients the prevalence of metastatic cancer pain may exceed 75%[1]. In patients with metastatic cancer of the prostate and breast, the prevalence of pain can reach 70% and is most often seen in patients with disease involving the spine. Symptoms of sharp, radiating pain are extremely common in these patients and are likely to be mediated by substances secreted by these tumor cells that affect neighboring spinal nerve roots [2].

Complicating the picture in patients with advanced prostate cancer is the fact that many patients are elderly, and therefore functionally limited, but may survive months or years in severe pain before succumbing to their disease. These patients often require very large doses of either systemic or intraspinal opioids that provide incomplete pain relief and that frequently produce undesirable side effects [2]. Clearly, the identification of tumor-associated mediators and their receptors that cause pain in metastatic prostate or other cancers could lead to the development of more effective treatments for this complication (Figure 1). One candidate mediator, the potent vasoconstrictor peptide, endothelin-1 (ET-1), [3,4] is secreted in high concentrations by metastatic prostate and breast cancer cells [5] and could cause spinal nerve root pain by a direct effect on nociceptive afferents (see below).

Figure 1

Lateral view of the L4 and L5 lumbar vertebral bodies showing an expanding tumor metastasis growing adjacent to and impinging upon the L4 spinal nerve root. Several candidate mediators that could either directly stimulate local pain fibers in nerve root and surrounding tissues, or that could sensitize these fibers to noxious or non-noxious stimuli are shown.

Endothelin-1 as a Pain Mediator

Endothelin-1 is a highly potent vasoconstrictor that was discovered by Hickey et al. [3] and then isolated, sequenced and cloned by Yanagisawa et al. [4]. Most efforts to establish the function of ET-1 have examined its potential role in cardiovascular disease. However, recent evidence is accumulating to suggest that ET-1 may contribute to a variety of pain states in animals and in humans [6–12]. In support of these observations, ET-1 is present in high concentration in both dorsal root ganglion (DRG) neurons [13] and satellite cells, [14] where it may have paracrine or autocrine functions [15]. In rodents, intraperitoneal administration of ET-1 induces abdominal writhing behavior that is ET-1 receptor-mediated [16,17]. Intra-articular administration of ET-1 has also been shown to induce pain-like behavior in rodents that is ET-1 receptor-mediated, while ET-1 can potentiate pain states in animal models of acute chemical- or inflammation-induced pain [6,7,10]. In humans, the injection of ET-1 into the brachial artery was reported to induce severe pain and prolonged, touch-evoked allodynia in the injected limb [18]. More recently, Carducci et al. [12] have reported that an endothelin-A receptor antagonist can reduce verbal reports of pain in patients with metastatic prostate cancer.

In support of these findings in animals and in humans, we have recently described pain behaviors in the rat that develop shortly after the extraneural or intraneural application of ET-1 to sciatic nerve, or after subcutaneous injection into the plantar hindpaw [7,19,20].

Mechanisms of Endothelin-1–induced Pain

Although the mechanism of action for ET-1 induced pain in animals or humans has not been established, our recent studies suggest that this pain may be due to the direct excitation of sensory fibers [9,20]. For example, we have observed that the application of ET-1 to rat sciatic nerve induces pain behavior, whereas identical application of equipotent vasoconstrictors (Zochodone et al., '92), such as epinephrine, does not induce pain behavior [9]. Furthermore, we have recently shown that subcutaneous injection of ET-1 into the plantar hindpaw of the rat both induces pain behavior and, quickly and selectively, excites C and Aδ (pain) fibers, but not Aβ (non-pain) fibers[20] (Figure 2).

Figure 2

Schematic of results obtained from subcutaneous administration of endothelin-1 to rat plantar hindpaw demonstrating selective excitation of pain fibers. The “naked nerve ending” receptor of a C fiber and the specialized receptor (Pacinian corpuscle) of a touch (Aβ) fiber are shown as well.

At the cellular level, pain and excitation of sensory fibers by ET-1 is most likely mediated by G-protein–coupled ET receptors linked to downstream signal transduction pathways that can activate ionic currents to depolarize sensory neurons. Two different mammalian G-protein–coupled ET receptors (ETA and ETB) that mediate the actions of ET-1 have recently been cloned and their pharmacology studied in a variety of tissues that include vascular smooth muscle [22], vascular endothelium [23], and the central [24,25] and peripheral nervous systems [13,21,26]. In smooth muscle cell (SMC) preparations, where ETA receptors are believed to mediate the vasoconstrictive actions of ET-1, numerous studies have demonstrated that ET-1 administration induces increases in intracellular Ca2+[23]. This increase is due to both an initial transient mobilization of intracellular Ca2+ stores [27,28], followed by a more sustained influx of extracellular Ca2+ through voltage-sensitive L-type Ca2+ channels, likely via mechanisms that involve phospholipase C activation and IP3-mediated activation of Ca2+-dependent chloride channels, respectively [23]. In brain vascular endothelium, ET-1 activates ETA receptors leading to inhibition of adenylyl cyclase via pertussis toxin-insensitive pathways [29]. In microvessels of the central and peripheral nervous system, locally applied ET-1 causes vasospasm and reduced blood flow [21,30,31] that is mediated in brain microvessels by ETA receptors [23].

Although the receptor dependence and transduction pathways that mediate the effects of ET-1 on sensory neurons have not yet been established, recent studies have demonstrated that ET-1– and inflammation-induced articular pain in rats is ET receptor-dependent [10]. ET-1 has also been reported to enhance capsaicin-induced pain in mice [32], and to induce peptide release and cGMP accumulation in sensory neurons [33]. Moreover, our recent studies have demonstrated that pain behavior induced by ET-1 applied to rat sciatic nerve or skin is both ETA receptor-dependent, and likely due to selective actions of ET-1 on nociceptive primary afferents [9,19,20](Figure 3). Consistent with these observations, it has been recently reported that ETA receptors are found both on primary afferent axons and on the cell bodies of small-diameter dorsal root ganglion neurons, which are believed to subserve pain sensation ( J. Pomonis and P. Mantyh, personal communication, October 2000).

Figure 3

Illustration of an axonal nerve cell membrane and a Schwann cell from rat sciatic nerve suggesting the likely site of action of extra- or intraneurally applied ET-1 on endothelin-A (ETA) receptors of pain fibers. ETB receptors are believed to be present on Schwann cells (J. Pomonis and P. Mantyh, personal communication) and do not appear to contribute to pain behavior produced by application of ET-1 to rat sciatic nerve [9] (Fareed et al., 2000).

Additional studies examining the effect of ET-1 on neuronal cells in culture may help us to establish the ionic and cellular basis for ET-1's pain-inducing actions. These cellular and molecular targets can subsequently be examined for their therapeutic importance in the treatment of ET-1-associated pain in preclinical models and in humans.


Metastatic cancer pain remains a challenging clinical problem, the cause of which remains uncertain. However, recent studies provide evidence that chemical mediators made by cancer cells may contribute to this pain. Endothelin-1 shows particular promise as a candidate mediator of cancer pain since it is capable of causing pain in animals and in humans that, in animals, is receptor-dependent and probably due to selective actions on pain fibers. Establishing a role for tumor mediators like endothelin-1 in humans with metastatic cancer pain may allow us to directly target the cause of this pain, without producing undesirable side effects (e.g., excess sedation) that are often associated with broader spectrum (e.g., opioid) analgesics.


The author gratefully acknowledges the contributions of his co-investigators. Supported in part by funds from USPHS grant CA 80153.


View Abstract