Print this page    •   Back to Web version of article

Experimental Treatments for Kaposi's Sarcoma

By Michael Marco and Martin Majchrowicz

July 25, 1994

There are more experimental drugs for KS in clinical trials in 1994 than ever before. Some trials are looking at IFN-alpha and cytotoxic agents -- usually in combination with AZT, ddI and ddC. Others are more novel approaches, such as cytokine and immune modulators, angiogenesis inhibitors, currently approved cytotoxic agents used in other cancers, and liposomally encapsulated anthracyclines.

It is up to patients -- and their physicians -- to determine if they want to use standard regimens to treat their KS or volunteer for a clinical trial. Many of these compounds seem appealing and some have even shown activity. Patients, however, must realize that most of these compounds are in phase I or II of their development. Long term efficacy and toxicity findings are, of course, not available.

Liposomal Formulated Anthracyclines

DOX-SL (DOXIL) (Liposome Technology Inc.)

DOX-SL is a liposomal formulation of doxorubicin (Adriamycin). Unlike most conventional liposomes -- tiny spheres composed of durable and primarily biodegradable lipid membranes -- DOX-SL is a trademark "Stealth" liposome that is surrounded by long chains of polyethylene glycol (PEG). These PEG coated chains alter the surface of the liposomes, thus shielding them from phagocytosis (engulfment) by the immune system. Thus, these Stealth liposomes are able to travel through the bloodstream for a prolonged period of time and deliver their contents directly to the site of disease. It is hoped that this liposomal encapsulations will be able to deliver DOX-SL more effectively.

DOX-SL is currently in three separate trials for KS. LTI's randomized phase III trial (protocol # 30-10), DOX-SL versus ABV has accrued roughly 130 patients and another 100 are needed for adequate statistical power. Patients must have at least 25 mucocutaneous lesions and/or the development of 10 or more a month. Also, patients with just visceral disease and at least 2 assessable cutaneous lesions are also eligible. Patients who have had prior anthracycline therapy (doxorubicin) are not eligible.

DOX-SL is also available, through protocol #30-12, to patients who have failed or are intolerant of prior cytotoxic therapy. All patients who have completed six courses (at two week intervals) of DOX-SL or ABV in # 30-10, also roll-over into 30-12, where all will receive DOX-SL. To date, approximately 300 patients are in this trial.

Data from 50 or so patients from LTI's non-randomized, DOX-SL trial, 30-12, will be part of the DOX-SL NDA, to be submitted to the FDA in mid 1994, for approval in KS patients who have failed or are intolerant of standard cytotoxic therapy. LTI plans to go to the FDA to expand the initial, salvage indication, to include first-line chemotherapy with DOX-SL, after 30-10 is completed. LTI plans to close 30-12 to enrollment in mid-1994. Future salvage patients may be able to obtain DOX-SL through a planned Treatment IND, if the FDA permits this.

Results from DOX-SL phase I and II trials demonstrate that it is safe and yields higher response rates than those reported with single agent doxorubicin (as seen in ACTG 006 conducted by Fischl and colleagues) (Fischl 1993). Northfelt and colleagues (Northfelt 1993) in their phase I trial noted that DOX-SL has a half-life in the blood of over 40 hours compared to free doxorubicin, which only has a half-life of less than 5 minutes. Moreover, DOX-SL was present in KS lesions at concentrations 5 to 11 times that after the same dose of free doxorubicin, and was also present at 10 to 20 times higher concentration in KS lesions than in normal skin (Northfelt 1994).

Results of various open phase I/II and II/III trials were released at the IX International Conference on AIDS in Berlin. Response rates (ambiguously and subjectively defined) for DOX-SL, as high as 70%, were reported at doses of 10-20 mg/m2 given every two weeks (Goebel 1993; Jablonowski 1993). Many patients in these trials had progressive disease and had previously taken other cytotoxic agents.

Toxicities included mild to moderate neutropenia, alopecia and anemia. Most trials allowed for G-CSF which raised patients' ANC and allowed them to stay on drug. Cardiac toxicity, which has been seen with free doxorubicin at cumulative doses of over 500 mg/m2 in other cancers, was not noted in these trials (George Tidmarsh, LTI, personal communication). Because of toxicity, intercurrent illness or tumor progression, however, most KS patients do not stay on doxorubicin long enough to receive cumulative doses of over 500 mg/m2. Patients in the DOX-SL trials have not yet received cumulative doses near 500 mg/m2, so further monitoring will be warranted for those on maintenance.

Most recently, preliminary results on 48 patients from DOX-SL 30-12 indicated a 66.7% partial response rate (Thommes 1994). The mean duration of this partial PR was approximately 83 days.

If DOX-SL is approved, many researchers believe that oncologists will try DOX-SL in combination with BV (Jamie Von Roenn, personal communication). This is a logical substitution to standard ABV if DOX-SL's response rates (from the current trials) are higher and than those previously seen with doxorubicin and/or less toxic. Ronald Mitsuyasu, along with members from the KS Working Group of the ACTG Oncology Committee developed a concept sheet for a DOX-SL trial (DOX-SL vs. DOX-SL with BV) to ascertain how DOX-SL might be used most effectively (considering the toxicities) on chemotherapy naive patients.

Patients will get DOX-SL on either arm and will receive -- if necessary -- G-CSF to avoid neutropenia. This study also helps LTI because they only have to supply the drug and will not have to foot-the-bill for the extensive data monitoring. If the ACTG's Adult Review Group approves the concept sheet, this trial should be up and running by late 1994 (Jamie Von Roenn, personal communication).

Daunoxome (Vestar)

Daunoxome is liposomal-encapsulated daunorubicin. Unlike DOX-SL with its PEG coating, Daunoxome is a conventional liposome. Phase III randomized trials of Daunoxome versus ABV are open and actively recruiting patients at approximately 11 centers around the country. So far, approximately 200 patients have received drug and another 50 are needed (Michael Ross, Vestar, personal communication). Vestar also set up a salvage protocol, which unfortunately is ongoing at only 7 sites in major US cities, for patients who have failed on at least one cytotoxic agent.

A phase I pharmacokinetic study conducted by Gill (Gill 1991 b) revealed that Daunoxome has a half-life of approximately 2 hours compared with free daunorubicin which has a half-life of approximately 3 1/2 minutes. Data from their phase II trial was also delivered at the Berlin AIDS conference. From a 25 patient Phase II study of Daunoxome (40 mg/m2 every two weeks), Chew and colleagues (Chew 1993) reported: one complete response, 20 partial responses, and one with a stable disease. Toxicities noted were mild nausea and neutropenia. G-CSF was needed in six of the 25 patients to raise the ANC.

What the company believed to be promising response rates from this trial and others (Presant 1993), led Vestar to submit an NDA to the FDA for approval of Daunoxome for KS patients who are either refractory or intolerant to standard chemotherapy. Vestar's application, with data on approximately 50 patients, of whom 39 were deemed evaluable, was from open-label, uncontrolled, Daunoxome trials (40 mg/m2 every two weeks). The patients included in the data package were selected because they had failed or were intolerant of prior chemotherapy regimens (Michael Ross, Vestar, personal communication).

The FDA's Oncologic Drugs Advisory Committee on June 17, 1993 rejected Vestar's NDA on the basis of insufficient data (The Pink Sheet, 55: 21 June 1993). Vestar, in its analysis of the data, reported a 51% partial response rate. The FDA, in a separate analysis (but still using the protocol's response criteria), lowered the response rate to 18%. The discrepancy centered around definition of a partial response. Vestar's partial responses were often based on the "measure of flattening" instead of "reduction in the number or surface area." (Pink Sheet) Moreover, many of the patients had "breakthrough" lesions (i.e., developed new lesions as others regressed), or their responses did not last for the required 4 weeks.

Treatment activists and noted oncologists felt that both Vestar and the FDA handled the ensuing events poorly. Vestar should have had cleaner data and should have followed the response definitions. On the other hand, the FDA could have notified Vestar (during the 3+ months when they had the data) as to the discrepancies found in their analysis of the data. Vestar might not have come and the fiasco could have been avoided.

Angiogenesis Inhibitors

Angiogenesis -- the process of forming new blood vessels -- is a natural phenomenon that takes place during certain physiologic processes such as wound healing or during a woman's reproductive cycle. Outside those circumstances, angiogenesis can be pathological. Evidence now suggests that Kaposi's sarcoma might not be a "true" malignancy (a clonal proliferation of genetically altered cells), but a dysregulation of angiogenesis. In fact, spindle cells (KS cells) that have been cultured -- as well as injected into nude mice -- respond to and secrete a number of cytokines (growth factors) including basic fibroblast growth factor (bFGF), vascular endothelial cell growth factor (VEGF), platelet-derived growth factor (PDGF), granulocyte macrophage colony stimulating factor (GM-CSF), IL-1, and IL-6 (Ensoli 1989; Nakamura 1988). Some of these cytokines secreted by spindle cells are thought to induce a proliferation of endothelial cells (neovascularization) that lead to the poorly organized vascular formulations found in the initial stage of KS lesions.

While many researchers believe these cytokine and growth factor produce this neovascularization which leads to the formation of new KS lesions, not all agree as to their exact mechanisms of action. Robert Gallo and his colleagues contend that KS begins with an HIV-associated inflammatory process of activated T cells, which secrete IL-1, IL-6, TNF, GM-CSF and gamma Interferon. These cytokines, in turn, induce endothelial cells to proliferate and transform into spindle-shaped KS cells that comprise the initial KS lesions (RC Gallo, personal communication). The KS spindle cells secrete large amounts of bFGF, VEGF, and PDGF -- as well as becoming responsive to Tat -- which create an ongoing cycle allowing new lesions to appear. It is felt that angiogenesis is central to the pathogenesis of KS. As Sam Broder said, "The blood vessels are the tumor."

Because solid tumors need a blood supply to grow and spread, scientific research has looked into ways to "starve" solid tumors to prevent this abnormal blood vessel development. This angiogenesis research has lead to the development of a new class of compounds called angiogenesis inhibitors. Three such compounds, Tecogalan (SPPG, DS-4152), TNP 470, and recombinant human Platelet Factor 4 (rPF4) are currently in phase I safety and efficacy trials for patients with KS and various cancers.

These compounds have only been tested in a small number of patients with KS and have mostly induced a lessening of KS-associated edema and some minor activity on existing lesions. Many researchers feel that angiogenesis inhibitors might be most effective in preventing new lesions from occurring rather than affecting preexisting lesions (Pluda 1993 a). If this is true, these drugs might be optimal for maintenance therapy after systemic chemotherapy.

Tecogalan (SP-PG: Daiichi)

Tecogalan is an angiogenesis inhibitor that was isolated from a low molecular weight fraction of a sulfated polysaccharide that originates from the bacterium Arthrobacter. Tecogalan is thought to inhibit the binding of bFGF to the endothelial cell, thus inhibiting bFGF-stimulated migration and proliferation. Gallo and colleagues showed that Tecogalan, administered to "nude" mice (special immunodeficient mice) inoculated with AIDS-KS spindle cells, "led to a degeneration of newly formed vascular lesions." (Nakamura 1992).

Four sites are actively recruiting patients for a phase I dose-escalating pharmacokinetics and safety trials of intravenous Tecogalan. Thus far, no serious toxicity has been seen, except for prolonged APTT (activated partial thromboplastin time) which normalizes shortly after the end of infusion. Prolongation of the infusion time of a single dose of Tecogalan decreases the prolongation of APTT. Some patients also noted fever and chills after dosing, but such symptoms are ameliorated with meperidine (Demerol) and acetaminophen (Tylenol). Initial assessments of Tecogalan in the uncontrolled phase I trials suggest a decrease in KS-associated edema (swelling) and possible lesion reduction in some patients (Gill 1994).

There is no specified CD4 range for these trials, but patients must have at least 5 cutaneous lesions and no evidence of symptomatic visceral involvement. Sites require 24 hours of hospitalization for monitoring and pharmacokinetics during initial dosing. Subsequent doses will be given on an outpatient basis.

TNP-470 (AGM-1470: Takeda Chemical Ind., Ltd.)

TNP-470 is an analogue (a closely related chemical) of fumagillin which is a naturally secreted product of the fungus Aspergillus fumigatus fresenius. Fumagillin was originally found, in the 1940s, to have some effect against amoebas, but was toxic in animals. In 1990, Ingber and colleagues found that fumagillin and its synthetic analogue, TNP-470, exhibited potent activity against endothelial cells and inhibited the growth of solid tumors without side effects when injected into mice. TNP-470 was found to be less toxic in animals and 50 times more potent in vitro than its fumagillin parent (Ingber 1990). Yarchoan and colleagues at the National Cancer Institute (NCI) also found that TNP-470 inhibited spindle cells in KS cell lines (Robert Yarchoan, personal communication).

Administration of extremely high doses given by bolus injection every other day induced small hemorrhages in the brain, lungs, heart and retinas of dogs. However, the drug was better tolerated when given as an infusion. Hence, initial clinical testing favored a one hour infusion starting at low doses.

In preliminary results from an NCI phase I study of TNP-470, given intravenously for one hour every other day in a dose- escalating schema, Pluda and colleagues reported a lessening of edema, activity on existing tumors and a decrease in the development of new lesions with no dose-limiting toxicities (Pluda 1993 c).

This trial, like the four-site study by the ACTG (20 - 50 mg/m2 IV weekly), is actively recruiting patients. There is no CD4 level requirement for these dose-escalating trials. Patients with pulmonary KS are excluded from the NCI trial, and patients with pulmonary and/or symptomatic visceral KS are excluded from the ACTG trials.

Recombinant Platelet Factor 4 (rPF4: Repligen)

rPF4, purified from genetically engineered Escherichia coli, has been found to inhibit endothelial cell proliferation and migration (Maione 1990). Miles and colleagues found that rPF4 also inhibited the proliferation of cultured AIDS-KS cells in a dose dependent manner (Miles 1991).

In the initial phase I/II safety and preliminary efficacy trial, rPF4 was injected into one KS lesion while another control lesion was injected with saline. Six of the 12 patients injected with rPF4 demonstrated an anti-KS effect, the only toxicity being mild erythema (redness around the lesion) in approximately 20% of the patients (Kahn 1993). More recently, preliminary data was released from an ongoing intralesional rPF4 trial with seven evaluable patients (Staddon 1994). From the seven patients who had one lesion injected with rPF4, there were 2 CRs, 2PRs, 2 with stable disease and 1 with progressive disease. There were no CRs and only one PR noted in the non-injected, proximal and distant control lesions. Investigators noted that while the drug seems active, "intralesional dosing appears suboptimal to achieve systemic levels."

There are three Phase I/II trials currently underway and actively recruiting patients for rPF4 at different sites in the country. The trials differ on the how rPF4 is administered: intralesionally, subcutaneously, and intravenously. The duration of these dose escalating trials is a maximum of eight weeks to assess responses and toxicity. In the intravenous study, there are two arms: one for patients with CD4s below 200 and another for those with over 200 CD4s. Patients with cutaneous and visceral KS are eligible. The intralesional study excludes those with pulmonary KS. Patients receiving prior systemic chemotherapy are excluded from the subcutaneous study.

Cytokine Modulation

Recent insight into the role of cytokines in the pathogenesis of KS lesions has suggested that inhibitory strategies, aimed at inhibiting the production or action of cytokines, might be used to treat KS. As with the angiogenesis inhibitors, the various cytokine inhibitors now in phase I and II trials have been tested on very few KS patients. Almost all of the published articles on their supposed effectiveness and possible mechanisms of action are from in vitro data from various KS cell lines.

Tumor Necrosis Factor (TNF) Inhibitors

Over the past few years there has been a great deal of discussion surrounding TNF's role in the pathogenesis of primary HIV infection. Elevated serum TNF-alpha levels increase with advancing stage of HIV infection (Lahdevirta 1988). TNF-alpha has also been demonstrated to enhance HIV replication in chronically infected T-cell and macrophage cell lines when administered alone (Wright 1988).

Many researchers also believe that blocking TNF or its effects might inhibit the development of KS. TNF not only causes the proliferation of KS cells, but it also increases IL-6 levels and alters its expression in normal endothelial cells (Parkash Gill, personal communication). There are in fact, at least three different ways of blocking and or interfering with TNF: 1) by using a drugs such as pentoxifylline, thalidomide (and maybe Tenidap, made by Pfizer) that "decrease the production" of TNF; 2) by using a monoclonal antibody to "neutralize" TNF; 3) modulating the circulating TNF by "blocking" it from the receptor of the cell (Steven A. Miles, personal communication).

Pentoxifylline (Trental: Hoechst-Roussel)

Pentoxifylline (Trental), a methylxanthine (caffeine) derivative is an FDA approved drug used to treat intermittent claudication.6 When used in cancer patients, Dezube and colleagues found that pentoxifylline decreased TNF-alpha RNA levels in monocytes (Dezube 1990). Pentoxifylline has also been found to inhibit HIV-1 replication in infected human peripheral blood mononuclear cells (Fazely 1991).

Because pentoxifylline was found to decrease production of TNF, it was studied in KS patients in a phase I/II dose-escalating trial. In a UCLA trial that enrolled 12 patients (6 patients with a CD4 count above 200 and 6 patients with CD4 counts below 200), there were 3 responder; 1 CR and 2 PRs (Ronald Mitsuyasu, personal communication). These were from the over 200 CD4 group. Investigators do not believe that Hoechst-Roussel has any further plans for its clinical development in KS.

Recombinant Soluble TNF Receptor (r hu TNFr:Fc : Immunex)

Soluble TNF receptors might be beneficial by blocking TNF from binding to TNF receptors on KS spindle cells. rhu TNFr:Fc is a dimer of two molecules of the p80 TNF receptor linked by the Fc portion of immunoglobulin (Dower 1990). It binds TNF alpha and beta with high affinity. While it has no direct effect on KS cells, it is believed that it will block the effects of TNF, because the soluble receptor will act as decoy receptor and may prevent TNF from binding to the cell.

Miles and colleagues at UCLA are beginning a phase II placebo-controlled trial of rhu TNFr:Fc for KS patients. Walker and colleagues at the NIH are using rhu TNFr:Fc in a phase I/II trial in patients with HIV. In the UCLA 12 week safety and efficacy trial, patients will be randomized to receive either 125 micrograms/m2, subcutaneously twice weekly for 12 weeks or a placebo. Up to 40 patients will enroll. At least half of the enrolled patients will have a CD4 count above 200. The patients with over 200 CD4 cells will not be eligible if they have had a prior OI and the patients with below 200 CD4 cells will not be eligible if they have uncontrolled concurrent OIs.

Response rates with the soluble TNF receptor might only be about 30% (as was noted in the UCLA pentoxifylline study). This may be because only a third of KS patients probably have a TNF-driven mechanism (Steven Miles, personal communication). However, the TNF receptor might also prove beneficial for other HIV-associated conditions such as wasting syndrome.

Interleukin-4 (IL-4, Schering-Plough)

Studies have shown that interleukin-4 (IL-4) is a potent inhibitor of IL-6 secretion by monocytes in vitro (te Velde 1990; Miles 1992). IL-4 has also been shown to suppress the growth of myeloma cells in vitro by blocking endogenous IL-6 production (Herrmann 1991). Tepper and colleagues (Tepper 1989) conducted studies of malignant cells transfected with IL-4 cDNA that were introduced into tumor-bearing mice. Growth inhibition was noted in a variety of tumor types, including plasmacytoma, mammary adenocarcinoma, transformed fibroblast cell lines, a melanoma and sarcoma cell line.

The only toxicity seen in monkeys was cardiac toxicity. At extremely high doses (50 micrograms per kilogram per day), all ten monkeys died of multifocal myocardial necrosis. However, when monkeys were given 5 micrograms per kilogram per day, no cardiac toxicity was seen.

An ACTG dose-escalating phase I/II trial of recombinant IL-4 is in progress and actively recruiting KS patients. ACTG trial 224 is currently being conducted at UCLA with Steve Miles and at New England Deaconess Medical Center with David Scadden. A separate trial of IL-4 at USC is being conducted by Parkash Gill and has a few spots left for patients. All patients on this study must be receiving some form of antiviral therapy for at least 21 days prior to starting treatment. Patients with active OIs will be excluded. Group 1 will have under 100 CD4 cells. If they tolerate the treatment for 12 weeks, then group 2, with CD4 counts from 100-500, will start treatment.

Giving IL-4 to HIV-positive patients was controversial when this study was being planned because some believed that IL-4 might increase viral load. Researchers such as Gene Shearer of the NCI told Steve Miles that IL-4 was probably the worst thing you could give to an HIV-positive patient (Steven A. Miles, personal communication), because IL-4 is thought to shift the balance of CD4 cells from TH1 type (cytotoxic) towards TH2 type (immunoglobulin producing).

Now that IL-4 has been given to a handful a patients in both the ACTG trial and 17 patients in the USC trial, just the opposite has occurred. IL-4 is has not show much activity on KS lesions. There has been only 1 PR in the UCLA trial and only 1 PR in the USC trial. (Steven A. Miles, personal communication; Parkash Gill, personal communication). However, Miles has seen "significant decreases" in patient's p24 antigen levels. Decreases in patients' p24 levels have also been seen by Parkash Gill and by Lawrence Kaplan, who is using IL-4 in a lymphoma trial.

Soluble Interleukin-1 Receptor (IL-1r, rhu IL-1R: Immunex)

IL-1 has been shown to play a major role in the pathogenesis of HIV (Fauci 1993) as well as HIV associated KS. IL-1, a pro-inflammatory cytokine, acts as an autocrine growth factor for smooth muscle cells and induces the secretion of Platelet derived growth factor (PDGF) (Dinarello 1991). IL-1 has also been shown to increase production of TNF and IL-6 from peripheral blood mononuclear cells (Granowitz 1992). In vitro, IL-1 has been shown to stimulate the growth of KS cells (Nakamura 1988). It has thus become evident that by blocking IL-1, you might be able to stop its autocrine effects as well its effect on increasing IL-6 secretion. One possible way of blocking IL-1 is with the use of a soluble receptor that would act as a decoy receptor and prevent both IL-1 alpha and beta from binding to the surface of the cell.

rhu IL-1R is the soluble extracellular portion of the human type I IL-1 receptor (Dower 1989). It binds tightly to IL-1 beta and IL-1 alpha. While it has no effect on basal proliferation rates of KS cells, it can completely abrogate the IL-1 specific induction of proliferation (Steven A. Miles, personal communication).

Susan Krown at Memorial Sloan Kettering Cancer Center in New York has begun a phase I/II safety and efficacy trial of a soluble recombinant human IL-1 receptor for HIV positive patients with or without KS. This 8 week trial of subcutaneously administered soluble rhu IL-1R is currently recruiting patients with less than 300 CD4 cells.

Interleukin-2 (IL-2, Leukine, Chiron)

Intravenous IL-2 is being studied for KS in a small phase II trial at the NIH. This trial is part of the larger continuous IL-2 infusion trial that is currently under way at the NIH in HIV-positive patients.

Part of the rationale for testing IL-2 in KS patients resulted from the dramatic increase in CD4 cells seen by Kovacs and colleagues (Kovacs 1993) in their 8 patient trial of IL-2. Kovacs hopes IL-2 will show activity in patients with KS lesions who are also at risk for other OIs (Joseph Kovacs, personal communication).

This trial, with a target accrual of ten patients, has two arms: one for patients with over 200 CD4 cells and another for between 100 and 200 CD4 cells. Patients will receive IL-2 as a continuous infusion (18 million units for five days) as well as AZT, ddI, or ddC. Patients need to be on an antiretroviral because of the risk that IL-2, by stimulating T cell proliferation, will increase viral load.

Roswell Park Cancer Center is also running a phase II trial of IL-2. In this trial, for patients with KS and non-Hodgkin's lymphoma, IL-2 is administered at low doses by injection every day for one to three months.

CD8 Cell Expansion + IL-2 (Applied Immune Sciences)

CD8 cell expansion with the infusion of IL-2 has shown a hint of some anti-KS activity in clinical trials conducted by Klimas and colleagues (Klimas 1992, 1993). Two patients who had KS in a six-person phase I trial (Klimas 1992) had a response to therapy. One patient had a 70% regression of tumor size and number of lesions. The other patient had a partial resolution of oral and facial lesions but no change in his visceral lesions.

In a later KS-specific trial of 10 patients (6 had completed treatment and were evaluable), Klimas and colleagues (Klimas 1993) reported 5 partial responses. (Responses, however, were defined "by an overall reduction of greater than or equal to 25% in tumor load.")

In a separate analysis of that trial (with 8 patients who had completed treatment and were evaluable), AIS, using ACTG response definitions, claims there was 1 CR and 3 PRs (Annemarie Moseley, AIS, personal communication). The only toxicities noted by investigators were "flu-like" symptoms attributed to IL-2.

Kahn and colleagues completed a similar trial of CD8 expansion with IL-2 and recorded only 1 CR. AIS said they believed the reason for the discrepancy in the results between the two identical trials was: Klimas' patients had lower CD4 cell counts (between 50 and 150). Hence, CD8 expansion with IL-2 works better on KS patients with a less intact immune system (Annemarie Moseley, AIS, personal communication).

CD8 expansion is not being studied in any current KS trials. AIS says that they want to finish compiling the Klimas and Kahn data before decided what KS patient population they should study.

All Trans-Retinoic Acid

Retinoids -- derivatives of vitamin A -- are a class of drugs that have been widely used to treat various skin conditions. They also have immunomodulatory effects, including enhancement of lymphocyte and macrophage functions (Pluda 1993 a).

Warrell and colleges (Warrell 1991) reported clinical activity when all-trans retinoic acid was administered to patients with acute promyelocytic leukemia. All-trans retinoic acid has also exhibited anti-angiogenic activity in the chick embryo chorioallantoic membrane assay system (Oikawa 1989). Additionally, Sidell and colleagues (Sidell 1991) have found that retinoic acid inhibited the growth of a myeloma cell in vitro via down-regulation of IL-6 receptors. Gill and colleagues, however, found that this might not be the case in KS (Parkash Gill, personal communication). They have found that TRA down-regulates IL-6 (not the receptor) in vitro.

In an oral TRA trial (2 mg/kg) of seven patients conducted by Bonhomme and colleagues (Bonhomme 1993), one patient had a "significant response," two had a partial response, three remained stable, and one progressed. Duration of response was short, with patients experiencing new lesions shortly after discontinuing treatment. Toxicities included dry skin, myalgia, and headaches.

Disconcerting results were seen in a 14 patient trial of oral TRA conducted by Von Roenn and colleagues (Von Roenn 1993). Of the first 6 patients who received a high dose of TRA 175 mg/m2 every day for approximately 32 days, 5 showed rapid KS progression. Rapid progression was not seen at lower doses. There could be a differential, dose-dependent effect on KS (Pluda 1993 a).

In a trial with topical TRA in eight KS patients with less than 100 CD4 cells, Bonhomme and colleagues (Bonhomme 1991) noted partial remission in seven patients. Clinical responses were based on change of color, and size of the lesions.

Presently, there are four different oral TRA trials -- two as a single agent and two dose-escalating in combination with varying doses of IFN-alpha. Two of the trials are closed and two others (one with Robert Yarchoan at the NCI and one with Susan Krown at Memorial Sloan-Kettering) are still open and accruing patients.

Gill and colleagues have seen a number of partial responses in their dose escalating trial (Parkash Gill, personal communication). Gill mentioned that it took a mean time of 22 weeks to actually see the responses. In fact, a few patient initially had mild exacerbations of their lesions, but went on to respond many weeks later. Moreover, the majority of the patients who responded had fewer than 100 CD4 cells. The toxicities he has noticed were mild, including headaches, ringing in the ears, dry skin and elevated triglyceride levels.

Another retinoid that might work on KS, 13-cis retinoic acid, in a topical and oral formulation, has been developed by Ligand Pharmaceuticals. The UCLA group is planning to start a trial in the next few months (Steven Miles, personal communication). This retinoid binds to different retinoid receptors than TRA.

Taxol (Paclitaxel: Bristol-Myers Squibb)

Taxol is a drug derived from the bark and needles of the pacific yew tree. Recently approved by the FDA for breast cancer, Taxol has also shown anti-cancer activity in ovarian and non-small cell lung cancer. In a lung cancer study by Murphy and colleagues (Murphy 1993), an overall 24% response rate was noted with one patient achieving a complete response. Some of the toxicities noticed in other trials, included mild neutropenia, anemia, alopecia and peripheral neuropathy (Taxol, package insert).

Taxol is also being tested in a dose-escalating phase II trial for KS patients at the NCI. The trial is open and currently recruiting patients with KS. Patients receive Taxol over a three hour infusion once every three weeks. Patients with symptomatic pulmonary KS and those who have previously received more than one cycle of cytotoxic chemotherapy are excluded.

Preliminary results, according to Robert Yarchoan, principal investigator for the study, demonstrate that Taxol is causing lesion reduction (Robert Yarchoan, personal communication). Of 8 evaluable patients who have received Taxol at low doses, 5 have achieved partial responses. Alopecia, mild fatigue, rash and chills have been the observed toxicities.


The number of advances in the pathogenesis of KS led to the development of compounds which had promise of antitumor activity. Although some compounds seem to work in vitro, they have shown no antitumor effects in KS patients, and they have been toxic. Below, are two compounds that have been studied in KS clinical trials without evidence of efficacy. While some of these compounds might be resurrected for other HIV complications, they are no longer being studied specifically for KS.

TAT Inhibitor (Hoffmann-LaRoche)

The Tat protein, made by HIV inside of infected cells, has been shown to increase the proliferation of KS spindle cells (Ensoli 1990), and to upregulate TNF expression.Tat proteins are secreted into the bloodstream and apparently bind to and activate KS spindle cells, forming an ongoing cycle for the proliferation of KS lesions (Robert C. Gallo, personal communication). Vogel and colleagues demonstrated that HIV Tat, under the control of the HIV long terminal repeat, could induce KS-like lesions in nude mice when transfected into mouse embryo cells (Vogel 1988).

A trial was designed for KS patients using Hoffmann-LaRoche's Tat inhibitor (Ro-24-7249). This was the same Tat inhibitor used in the HIV trials (ACTG 213) that proved, after months of speculation and hype, to be ineffective, and possibly harmful. Two centers enrolled 22 patients at the 600 mg dose. There was an exacerbation of KS lesions in some patients, and a great deal of central nervous system toxicity was seen (Ronald Mitsuyasu, personal communication).

Pentosan Polysulfate

Pentosan polysulfate has been shown to inhibit bFGF (basic fibroblast growth factor) and FGF-like dependent tumor growth in vitro and in vivo (Wellstein 1990, 1991). Pentosan polysulfate was also found to inhibit the formation of tumors in nude mice and inhibit the growth if Kaposi's sarcoma spindle cells in vitro (Nakamura 1992).

These findings led researchers to believe that pentosan polysulfate might work as an angiogenesis inhibitor for patients with KS. In a 16 patient phase I trial of pentosan polysulfate, given systemically or intralesionally, conducted by Pluda and colleagues (Pluda 1993 b) at the NCI, the investigator noted that no "significant antitumor responses" were observed in any of the patients.

Previous | Next
Table of Contents

This article was provided by Treatment Action Group. It is a part of the publication The KS Project Report: Current Issues in Research & Treatment of Kaposi's Sarcoma. You can find this article online by typing this address into your Web browser:

General Disclaimer: is designed for educational purposes only and is not engaged in rendering medical advice or professional services. The information provided through should not be used for diagnosing or treating a health problem or a disease. It is not a substitute for professional care. If you have or suspect you may have a health problem, consult your health care provider.