Investigational New Drugs 20: 209–219, 2002.
`© 2002 Kluwer Academic Publishers. Printed in the Netherlands.
`
`209
`
`Suramin’s development: what did we learn?
`
`Maninderjeet Kaur1, Eddie Reed2, Oliver Sartor3, William Dahut1 and William D. Figg1
`1Molecular Pharmacology Section, Cancer Therapeutic Branch, Center for Cancer Research, National Cancer In-
`stitute, National Institute of Health, Bethesda, MD; 2Mary Babb Randolph Cancer Center, West Virginia University,
`Morgantown, WV; 3Stanley Scott Cancer Center, Louisiana State University, New Orleans, LA, USA
`
`Key words: prostate cancer, suramin
`
`Summary
`
`Suramin, a polysulphonated napthylurea, has been extensively evaluated over the past 10 years as an anticancer
`agent, with the most interest in the treatment of prostate cancer. Early clinical results were promising with response
`rates of up to 70% being reported. However, a recent double-blind study showed only modest palliative effect in
`patients with androgen independent prostate cancer. In retrospect, it appears those initial reports failed to control
`for confounding variables such as antiandrogen withdrawal and hydrocortisone.
`Suramin causes numerous reversible toxicities (lethargy, rash, fatigue, anemia, hyperglycemia, hypocalcemia,
`coagulopathies, neutropenia, renal and hepatic complications). Neurotoxicity has been the most significant com-
`plication and appears to be related to the intensity of the dosing regimen. An optimal therapeutic dose has not been
`determined, but it is clear that adaptive controls add little benefit.
`Aside from moderate toxicities and the low therapeutic index in patients with prostate cancer, suramin’s
`development has taught us some valuable lessons (i.e., anti-androgen withdrawal was noted during suramin’s
`development, the use of PSA as an indicator of tumor burden was initiated during the evaluation of suramin).
`These lessons can be applied to all clinical trials in hormone refractory prostate cancer. Suramin has significantly
`enhanced the evolution of our knowledge in several areas of prostate cancer biology and treatment.
`
`Introduction
`
`Suramin, a polysulphonated napthylurea, was first
`synthesized in 1916 by Bayer AG [1]. It was noted to
`have trypanocidal activity and thus became the drug of
`choice for African trypanosomas and onchocerchiasis
`[2,3]. In 1979, suramin was noted to inhibit reverse
`transcriptase and thus was evaluated in various clin-
`ical trials in patients with Acquired Immunodeficiency
`Syndrome (AIDS) [4,5]. Although these clinical trials
`showed minimal activity against human immunodefi-
`ciency virus (HIV), the drug seemed promising in HIV
`associated neoplasms such as: Kaposi’s Sarcoma and
`Non-Hodgkins lymphoma [3]. On a molecular level,
`suramin has the ability to block the actions of vari-
`ous growth factors such as; fibroblast growth factors
`(FGF), platelet derived growth factors (PDGF), trans-
`forming growth factors alpha and beta (TGF), and
`
`insulin like growth factor I (IGFI) [6,7,8,9]. Suramin
`has also been shown to be a strong inhibitor of an-
`giogenesis. Other biological activities of suramin are
`similar to polyanionic glycosaminoglycans and hepar-
`inioids [7]. Preclinical data showed in vitro antipro-
`liferative activity against human prostate cancer cell
`lines (LNCaP, PC-3, and DU145) [8–10]. Therefore,
`suramin was evaluated for antineoplastic properties
`with emphasis in patients with prostate cancer.
`
`Suramin as a single agent in metastatic prostate
`cancer
`
`Initial trials used a continuous infusion of suramin
`and targeted concentration between 100–350 µg/ml
`[8,9]. Subsequently,
`it was felt that concentration
`below 100 µg/ml showed no significant biological
`
`002013
`
`AVENTIS EXHIBIT 2020
`Mylan v. Aventis, IPR2016-00712
`
`

`
`210
`
`activity and above 350 µg/ml significant neurotoxicity
`was reported [9,10]. In addition, increased inter- and
`intra-patient pharmacokinetic variability was observed
`which resulted in several studies using adaptive con-
`trol [9,10]. At this point, preliminary data suggested
`meaningful activity against prostate cancer.
`In 1992, a phase II trial conducted at the NCI
`evaluated efficacy and dosing schedule of suramin
`[8]. This trial enrolled 38 patients with androgen-
`independent prostate cancer (AIPC) and used con-
`tinuous infusion to reach a peak concentration of
`300 µg/ml at the end of two weeks. An eight week
`wash out period was given to recover from toxicity.
`Subsequent cycles were repeated. In 17 patients with
`measurable soft tissue disease, three patients exhibited
`complete response, three showed partial response, and
`five patients had prostate specific antigen (PSA) de-
`cline of 75% or more [11]. In the remaining 21 patients
`with bone metastasis, only eight patients showed PSA
`decreases of ≥ 75% and PSA levels normalized (<
`4.0 ng/ml) in five other patients. Figure 1 shows the
`difference in survival between those patients that had
`at least a 50% decline in PSA and those that did not.
`A four week landmark analysis revealed that the me-
`dian survival was 21.1 months for those patients with
`a ≥ 50% decline in PSA compared with 6.8 months
`for those patients with a < 50% decline in PSA.
`Subsequently, a phase I trial at University of
`Maryland Cancer Center was conducted which util-
`ized short intermittent infusion using adaptive control
`with feedback to maintain plasma drug concentrations
`between 150–250, 175–275, or 200–300 µg/ml in
`three different cohorts [12]. These investigators repor-
`ted that 77% of patients had 50% or more decline in
`PSA and 55% of patients had a reduction of 75% or
`more in PSA. Eighty three percent (83%) of patients
`experienced reduction in pain. The method of dosing
`suramin utilizing adaptive control is cumbersome and
`time consuming and was not clinically feasible on a
`large scale [12]. Therefore, Kobayashi and colleagues
`evaluated the activity of intermittent dosing of suramin
`using a fixed dose regimen (N=63) [13]. They recom-
`mended that on the first day a dose of 1440 mg/m2 be
`used in future trials with gradual decrease in dosing
`on days 2, 8, and 9 of a 4 week cycle [13]. These
`studies stressed the need for determining the optimal
`drug concentration range and duration of therapy, and
`the role of required therapeutic monitoring [13].
`In 1995 Panchivan et al. analyzed five different tri-
`als of suramin in AIPC and concluded that the overall
`response was 55% in regards to PSA declines of 50%
`
`or more and 30% in regards to measurable disease
`[14]. This combined response rate is lower than initial
`reports, but still appeared promising.
`A more recent double-blind, randomized, phase III
`trial comparing suramin plus hydrocortisone (n=288
`patients) vs. placebo plus hydrocortisone (n=230 pa-
`tients) found a statistically significant palliative ad-
`vantage for those treated with suramin [15]. The palli-
`ative advantage was shown by decrease in pain and/or
`decrease in opiod analgesic use and prolongation of
`pain response. The toxicities in this study were mild
`to moderate and easily medically manageable. Rash
`was the most common toxicity followed by asthenia,
`edema, nausea and vomiting. In addition, the PSA de-
`clined in 32% of patients receiving suramin vs. 16%
`of patients in the placebo arm [15]. However, the sur-
`vival was similar in both study groups. Compared to
`initial high response rates reported with suramin, this
`study was disappointing. In fact, the low response rate
`coupled with no survival advantages, plus the prob-
`lems of toxicity resulted in the recommendation of not
`approving suramin for marketing by an FDA advisory
`committee.
`
`PSA: can we rely on it?
`
`PSA, 34 kilodalton protein, is found in prostatic tissue
`and in seminal fluid [16]. Serum PSA values reflect
`prostate volume in both malignant and benign tissue
`and has been an essential tool in the diagnosis of pro-
`state cancer [14,17,18]. In 1987 Ferro et al. utilized
`PSA as a marker of tumor burden in a phase II trial in
`AIPC patients [19]. Subsequently, PSA reduction of
`50% or more have been reported by several groups to
`be associated with survival advantage [11,20]. How-
`ever, Thalmann et al. in 1996 reported that suramin
`had a differentiating effect on tumor growth and PSA
`expression in LNCaP cells (androgen dependent cell
`line) as well as in C4-2 cells (androgen independent
`cell line grown in vitro) [17]. They suggested that PSA
`values might not be an appropriate end point in clinical
`trials using suramin therapy in AIPC because declines
`in PSA might not be associated with tumor regression.
`This type of observation was confirmed both clinic-
`ally and pre-clinically with other compounds, such as
`CAI [21,22]. However, it should be pointed out that
`the effects of suramin on PSA expression have been
`mixed; some investigators have reported no effect on
`PSA expression [23]. Eisenberger et al. demonstrated
`the need for prospective clinical data by examining
`
`

`
`211
`
`Figure 1. Survival of patients with ≥ 50% PSA decline versus less than 50% PSA decline at 4 week after suramin treatment.
`
`the significance of PSA changes and reported a thera-
`peutic benefit of suramin in patients with prostate
`cancer who had decline in PSA [20]. Likewise the data
`presented in Figure 1 confirms those observations.
`Nonetheless, when compared with Eisenberger’s data,
`Small and colleagues [15] found a higher percent of
`patients treated with suramin had a PSA decline, but
`no difference in survival when compared to placebo.
`
`Controlling for confounding variables
`
`Various other trials confirmed activity of suramin
`against prostate cancer with variable response rates
`(Table 1). In 1993 Scher and colleagues made a key
`observation about the antitumor activity of discon-
`tinuing flutamide (Table 2) [24]. This maneuver was
`confirmed by other groups to have activity in 20–
`30% of AIPC patients [25]. In addition, a survival
`benefit was shown to be associated with cessation of
`flutamide [26]. This observation led to speculation of
`
`the impact of confounding variables such as flutam-
`ide withdrawal and hydrocortisone coadministration
`on previously reported studies evaluating suramin’s
`activity.
`Often patients discontinued their antiandrogen im-
`mediately before starting suramin. Because of the
`adrenal ablation associated with suramin, replacement
`doses of hydrocortisone were necessary to prevent
`adrenal insufficiency. Many patients received higher
`than replacement doses of hydrocortisone in the tri-
`als for treatment of rashes, etc. Corticosteriods alone
`may have palliative or objective response in AIPC
`[27]. Tannock et al. [27,29] reported response rates
`between 13.6 and 38% with 7.5–10 mg/day of pred-
`nisone. Likewise, Sartor et al. found a 33% response
`rate with prednisone [30].
`In 1995 a group at the NCI prospectively evaluated
`the activity of suramin while controlling for both the
`hydrocortisone variable and flutamide withdrawal in
`54 patients with AIPC [31]. Figure 2 shows schema
`of patient distribution and response [31]. This trial re-
`
`

`
`212
`
`Figure 2. Schema, entry of patients and response (> 50% PSA decline) to hydrocortisone and suramin therapies. N: number; D/C: discontinue;
`RR: response rate; NE: not evaluable; MO: months.
`
`Table 1. Response rates with suramin in clinical trials
`
`Trials
`
`PSA % response rate
`≥ 50%
`≥ 75%
`
`Objective
`Response rate
`CR + PR
`
`Subjective
`Response rate
`(pain relief)
`
`Dosing
`
`Eisenberger et al. [33]
`
`24/31 (77%)
`
`17/31 (55%)
`
`6/12 (50%)
`
`24/31 (83%)
`
`Kobayshi et al. [13]
`Eisenberger et al. [33]
`
`9/12 (75%)
`45/67 (67%)
`
`4/12 (33%)
`NR
`
`1/7 (14%)
`7/18 (40%)
`
`NR
`18/37 (49%)
`
`Reyno et al. [53]
`Rosen et al. [27]
`
`24/36 (67%)
`16/32 (50%)
`
`18/36 (50%)
`7/32 (22%)
`
`0/4 (0%)
`2/20 (10%)
`
`12/23 (52%)
`22/32 (69%)
`
`Myers et al. [10]
`Garcia-Schürmann et al. [40]
`
`13/38 (34%)
`7/27 (26%)
`
`21/38 (55%)
`NR
`
`Dawson et al. [31]
`
`7/37 (19%)
`
`Kelly et al. [32]
`Bowden et al. [50]
`
`5/28 (18%)
`13/75 (17.3%)
`
`NR
`
`NR
`NR
`
`6/17 (35%)
`Bone scan
`2/27 (7.4%)
`NR
`
`0/28 (0%)
`NR
`
`15/21 (71%)
`1/10 (10%)
`
`NR
`
`NR
`NR
`
`Fixed, intermittent, and
`adaptive control
`Fixed and intermittent
`Fixed, intermittent, and
`adaptive control
`Fixed
`Fixed, intermittent, and
`adaptive control
`CIVI
`CIVI, fixed, and adaptive
`control
`Fixed, intermittent, and
`adaptive control
`Fixed and intermittent
`CIVI and adaptive control
`
`CR = complete response, and PR = partial response based on the National Prostatic Cancer Project (NPCP) criteria. NR = not reported. CIVI =
`continuous infusion
`
`

`
`Table 2. Flutamide withdrawal response rates alone and in combination with suramin
`
`Study authors
`
`Therapy
`
`No. of
`patients
`
`> 50% decline in PSA
`
`Response
`
`213
`
`Median response
`duration (mo.)
`
`Figg WD et al. [25]
`Scher HI et al. [24]
`Srinivas S et al. [51]
`Herrada J et al. [52]
`
`FW
`FW
`FW
`FW
`
`Bowden C et al. [50]
`Dawson N et al. [31]
`
`FW+SUR
`FW+SUR
`
`21
`35
`82
`41
`
`7
`25
`
`FW = flutamide withdrawal; HC = hydrocoftisone; SUR = suramin
`
`ported response rate of 19% for suramin (based on a
`≥ 50% decline in PSA) with a median duration of 2.2
`months. Later that year Kelly et al. reported the results
`from a similar trial in which they controlled for flutam-
`ide withdrawal and hydrocortisone [32]. The response
`rate was 18%, based on ≥ 50% decline in PSA; no
`measurable response was reported. The low response
`rate of suramin was in agreement with Rosen et al.,
`who also controlled for flutamide withdrawal [27].
`In contrast, Eisenberger and colleagues retrospect-
`ively evaluated their database, controlling for flutam-
`ide withdrawal [33]. They concluded that no signi-
`ficant relationship existed between the PSA decline
`associated with response and prior flutamide with-
`drawal [33]. They reported a decline in PSA of 50%
`or more in 67% of their patients. Again, however,
`three prospective trials support the theory that simul-
`taneous flutamide withdrawal could account for much
`of suramin’s activity.
`
`Flutamide withdrawal and combination
`chemotherapy
`
`In another non-randomized trial, the combination of
`suramin plus aminoglutethimide was evaluated with
`or without simultaneous flutamide withdrawal [34].
`The authors noted a partial response rate of 14.2%
`when flutamide had been discontinued prior to ini-
`tiation of suramin, and aminoglutethimide, whereas
`a 44% partial response rate was reported for the co-
`hort in which suramin therapy was begun concomitant
`with antiandrogen withdrawal. Median survival (21.9
`vs. 14.2 months) and progression free survival at one
`
`Flutamide withdrawal alone
`4.7
`33.3%
`5
`29%
`3
`15%
`3.2
`28.2%
`Suramin at the time of flutamide withdrawal
`NA
`2.5
`
`85%
`44%
`
`year (27.1% vs. 19.8%) were also better in the simul-
`taneous antiandrogen withdrawal group. In addition,
`two other non-randomized observations support this
`hypothesis [35]. These data suggest that suramin plus
`aminoglutethimide increases the response rate asso-
`ciated with antiandrogen withdrawal; a theory which
`should be prospectively evaluated in future clinical
`trials.
`
`Activity in patients with androgen dependent
`disease
`
`Dawson et al. combined suramin (target plasma con-
`centration of 175–300 µg/ml) and hydrocortisone with
`leuprolide (an LHRH agonist) and flutamide in pre-
`viously untreated metastatic prostate cancer patients
`(N=50) [36]. An overall response rate of 67% was
`reported, combining complete and partial responses.
`Figure 3 depicts the overall survival in this popula-
`tion (update August 1999). The median survival was
`3.4 years. Although this was not a randomized trial,
`and the patient population evaluated had a poor pro-
`gnosis (D1 and D2 stage), this survival data seems
`promising. Nonetheless, Hussain et al. have recently
`reported on 62 previously untreated metastatic patients
`that receive combined androgen blockade plus four
`cycles of suramin [37]. The cycles were separated
`by 6 months and the treatment regimen was an ag-
`gressive fixed dose schedule. Fifty-four percent (54%)
`of patients had significant toxicity to suramin, with
`one drug toxicity related death. Furthermore, the re-
`sponse rate was not overly impressive (progression
`free survival was 14 months, CI 10 to 19 months)
`
`

`
`214
`
`and median survival time was 24 months [37]. At
`the time of publication, 68% of patients had failed
`therapy. This study concluded limited applicability
`for suramin in the treatment of patients with newly
`diagnosed metastatic prostate cancer.
`
`Survival data for suramin in AIPC
`
`Because of increased variability in the range of re-
`ported responses for single chemotherapeutic agents,
`Petrylak and colleagues investigated the effects of
`various prognostic factors on survival in patients with
`prostate cancer [38]. They concluded that serum LDH
`and elevation in alkaline phosphatase had the greatest
`impact as prognostic factors [38], whereas other stud-
`ies included performance status as the most important
`factor followed by bone metastasis, weight loss, and
`PSA level [39]. Suramin was compared with several
`single agent chemotherapies; no significant difference
`in survival was observed [38]. A more recent analysis
`by Garcia-Schürmann et al. suggested a survival ad-
`vantage for patients treated with suramin [40]. Our
`data at the NCI suggests that certain subsets of patients
`appear to benefit from suramin. Figure 4 shows the
`survival data of all AIPC patients treated with suramin
`at the NCI. The median survival is 13.5 months; 13%
`of the patients are alive 3 years later, and a few patients
`have still not progressed.
`However,
`the recent controlled trial discussed
`above, comparing suramin + hydrocortisone and hy-
`drocortisone + placebo in two groups found no overall
`survival advantage in the suramin group [15]. Another
`phase III trial measured urinary VEGF levels as a pre-
`dictor of response in therapy with suramin [41]. They
`report that VEGF decline of < 15% from baseline
`(pretreatment) was associated with increased survival.
`The authors recommend future investigation to find
`out whether this is an important biological activity of
`VEGF or simply an epiphenomenon [41].
`
`Clinical pharmacology
`
`The clinical pharmacological properties of suramin
`include: half-life of 50 days or more, renal clear-
`ance, high level of binding to plasma albumin (99%
`or more), and minimal metabolism [13]. Suramin ex-
`hibits poor oral absorption with high distribution to
`tissues, kidneys, and protein. It has been shown that
`whenever large doses of drug have been administered,
`
`sufficient drug crosses the blood brain barrier to affect
`neuronal metabolism [43]. Volume of distribution of
`this drug is 31–46 liters and 80% of the drug is ren-
`ally excreted [42]. Elimination half life of suramin in
`41–78 days [44]. After IV administration, the plasma
`concentration falls rapidly, then gradually with ter-
`minal half-life of approximately half days [45]. The
`prolonged elimination is due to excessive protein bind-
`ing and minimal metabolism [42]. Using 2 and 3
`compartment models, the half-life of suramin is 41
`days and 78 days, respectively [45]. Initially, trials
`at the NCI used continuous infusion and found sig-
`nificant toxicities [8]. It was believed that suramin
`exhibited marked interpatient variability and a nar-
`row therapeutic window [8]. Later, Eisenberger et al.
`use short bolus infusions of suramin, calculated on
`the basis of plasma concentrations of drug, popula-
`tion priors, and patients’ individual experience with
`drug (adaptive control with feedback) [33]. Adaptive
`control with feedback used a pharmacokinetic model
`fitted to individual patient dosing and concentration
`and guided the subsequent dosing. It was initially im-
`portant, but later recognized unnecessary to keep drug
`concentrations in a defined range because of small in-
`terpatient variability [13,32]. Kelly et al. encourage
`the use of empiric dosing without adaptive control
`[32]. This regimen reported lower toxicities and bet-
`ter tolerability than the NCI trial. Later, a UCLA trial
`evaluated intermittent dosing (interposing rest period
`between 2 cycles of 8 week therapy) with reasoning
`that long half-life and cumulative toxicity will allow
`for longer therapy and better results with fewer tox-
`icities [27]. The UCLA regimen was well tolerated
`[27]. Recently,
`intermittent bolus infusion of fixed
`dose have shown to be safe on the basis of toxicities,
`cost effectiveness, and have allowed for multi center
`outpatient trial [15].
`
`Toxicity
`
`Earlier studies reported a series of toxicities associated
`with suramin leading to discontinuation of treatment
`in many patients with AIPC, despite suramin’s thera-
`peutic promise [8,9,10,40]. Table 3 lists most of the
`toxicities associated with suramin. The majority are
`reversible in the first few weeks to months after dis-
`continuation of suramin. However, other potential
`irreversible toxicities, such as adrenal insufficiency,
`need careful monitoring not only during treatment but
`also in long-term follow-up [36]. Replacement therapy
`
`

`
`215
`
`Figure 3. An update Kaplan-Meier depicts survival in androgen dependent prostate cancer population.
`
`Figure 4. Kaplan-Meier survival curve of AIPC patients treated with suramin at the NCI. Median survival is 13.5 months.
`
`

`
`216
`
`Table 3. Reported toxicities with suramin
`
`Adrenal insufficiency
`Cardiac
`Arrythmias
`Chest pain
`Congestive heart failure
`Edema
`Constipation
`Dermatologic
`Electrolyte abnormalities
`Hyponatremia
`Hyperkalemia
`Hypocalcemia
`Hyperglycernia
`Hypophosphatemia
`Hypomagnesia
`Fever
`Hematologic
`Anemia
`Coagulopathies
`Lymphocytopenia
`Neutropenia
`Thromobocytopenia
`Hepatic
`Infection
`Lethargy/malaise
`Neurological toxicity
`Motor nerve neuropathy
`Sensorimeter neuropathy
`Guillain Barr´e Syndrome
`Paralysis
`Opthalmic
`Renal
`Acute renal failure
`BUN
`Elevated serum creatinine
`Renal failure
`
`+++
`- -
`- -
`- -
`!!
`+
`- -
`++
`+
`- -
`- -
`+
`++
`- -
`+
`+
`+++
`+++
`++
`+++
`+
`+
`+
`+
`∗∗
`∗∗
`
`/+++
`
`++
`+
`!!
`!!
`++
`+
`- -
`+
`+
`!!
`
`+++ = very high frequency, ++ = high frequency, + = moderate
`∗∗
`frequency, - - = low frequency; !! = rare case report,
`= appears to
`be dose limiting toxicity. (Refs: 8,12,27,31,33,34,36,40,45,50,53)
`
`of 30 mg/day of hydrocortisone has been administered
`during and after discontinuation of suramin. How-
`ever, a few patients have been able to discontinue
`replacement steroids several years after treatment.
`Preclinical and clinical data suggested the dose-
`limiting neurotoxicity might be linked to sustained
`plasma concentration of ≥ 350 µg/ml [8,12]. How-
`ever, Bitton et al. evaluated a continuous intravenous
`infusion trial and found significant grade 3 and grade
`4 neurotoxicity at plasma concentration of 275 µg/ml
`
`[45]. They recommended that the total suramin dose to
`be held at 157 mg/kg over a period of ≥ 8 weeks and
`to limit the period of exposure to plasma concentration
`greater than 200 µg/ml to ≤ 25 days [45]. Questions
`regarding optimal dosage and control of toxicity still
`remain unanswered.
`The most common dose-limiting toxicities repor-
`ted were a syndrome of malaise, fatigue, lethargy [33],
`neurotoxicity (paresthesia, motor weakness) [46], an-
`orexia, and keratoacanthoma and other dermatological
`complications [3]. Dose-limiting toxicities appear at
`higher doses and may surface even three months after
`treatment.
`Toxicity seems to be related to drug exposure (time
`above a particular concentration) [45]. Halabi et al. ex-
`amined the effects of three different fixed doses (low,
`intermediate, high) in 390 hormone refractory prostate
`cancer patients [47]. In this study, higher doses of
`suramin produced increased grade 3 and 4 toxicities
`[47]. Eisenberger’s group evaluated a pharmacokin-
`etically derived intermittent dosing trial – after four
`weeks of treatment, suramin was administered in a
`five day schedule to reach peak plasma concentration
`of 300, 350, and 400 µg/ml in three cohorts [48]. Pre-
`liminary data suggests that grade 3 and 4 dose-limiting
`toxicities increased with increased plasma concentra-
`tion, but suramin plasma concentrations of 300 µg/ml
`were reported to be active and well tolerated. They
`believed that an easier schedule of drug administra-
`tion would encourage the development of combination
`regimens and outpatient use of suramin [48].
`Two recent trials used an easier fixed schedule
`of suramin on an outpatient basis [15,37]. Hussain
`and colleagues used similar 78 day fixed regimen of
`suramin to maintain plasma concentrations between
`150–250 µg/ml, suramin treatment cycles were re-
`peated every six months for four cycles [37]. This trial
`reported significant broad spectrum toxicities, which
`led to the removal of 33 out of 59, 56% of newly dia-
`gnosed prostate cancer patients. Most common toxicit-
`ies, which led to the discontinuation of treatment, in
`order of significance, were neurotoxicity, cardiovascu-
`lar, weakness and fatigue. Even though neurotoxicity
`was the most common, researchers could not estab-
`lish any association between suramin peak and trough
`concentration and neurotoxicity. In general, toxicities
`increased with repeated cycles. In another randomized
`trial [15], Small et al. investigated 78 day fixed regi-
`men in hormone refractory prostate cancer population
`to maintain concentration between 100–300 µg/ml on
`outpatient basis [15]. Toxicities were classified as mild
`
`

`
`to moderate in intensity, and easily medically manage-
`able. Both the suramin and the placebo group reported
`higher toxicities than previous studies. Rash was the
`most common in the suramin group, and asthenia, ed-
`ema, nausea and vomiting were higher in both groups.
`Neurological, hepatic, and hematological toxicities
`were rare. On the basis of toxicity profile and ease of
`administration, authors recommend future trials with
`other agents of non-overlapping toxicity [15].
`
`Conclusion
`
`In earlier trials [8,27,31,33], suramin showed emer-
`ging anti-tumor activity. However, a recent double-
`blind study showed only modest palliative effects in
`AIPC patients [15]. Initial preclinical and clinical
`data of suramin as a treatment for prostate cancer
`showed promise with responses of up to 70% being
`reported [13,33]. These initial trials did not control
`for confounding variables such as antiandrogen with-
`drawal and hydrocortisone. Subsequent trials which
`controlled for these variables reported lower response
`rates [15,31]. Although some patients reported sub-
`jective improvement in bone pain, other endpoints
`were not improved (measurable disease respons or
`survival). Although the median survival in most of
`these trials is not impressive and controlled trials
`fail to detect survival advantage, there does appear
`to be a small subset of patients that have benefit-
`ted from suramin. One earlier hypothesis was that
`this prolonged benefit is a result of either the adrenal
`ablation associated with suramin or the prolonged ad-
`ministration of hydrocortisone. Recent double blind
`trials, which prospectively controlled for both the
`flutamide withdrawal and concomitant hydrocortisone
`administration [15], disproves this hypothesis. In ad-
`dition,
`there appears to be an additional cohort of
`patients with androgen dependent prostate cancer who
`have had prolonged responses to the administration
`of suramin concurrent with the initiation of androgen
`ablation [36]. However, a recent phase II trial [37]
`concludes that suramin plus hydrocortisone and an-
`drogen deprivation has limited applicability in newly
`diagnosed metastatic prostate cancer. Finally, the ad-
`ministration of suramin at the time of discontinuation
`of flutamide shows some promise (synergy similar to
`that being tested with CALGB 9583) and may warrant
`further evaluation.
`Suramin appears to cause numerous reversible tox-
`icities. Neurotoxicity has been the most significant
`
`217
`
`complication, but appears to be linked to those trials
`using more intensive treatment regimens. Significant
`broad spectrum toxicities were reported in the SWOG
`trial [37]. An optimal therapeutic dose has not yet been
`determined, but it is clear that adaptive control adds
`little benefit. After analyzing current trials, suramin
`does not seem to offer an advantage to patients with
`hormone refractory prostate cancer or hormone sens-
`itive prostate cancer. Besides moderate toxicities and
`the low therapeutic index in prostate cancer patients,
`suramin’s development has taught us some valuable
`lessons. First, the effect of anti-androgen withdrawal
`was not characterized when earlier studies were done.
`Second,
`the use of PSA as a surrogate marker of
`therapeutic benefit was starting and still continues
`to evolve. The finding that PSA concentration can
`fluctuate by various factors [23] (infection, current
`medications) independent of antitumor effect has been
`recently appreciated [49]. Third, the use of hydro-
`cortisone with suramin let to the finding that hydro-
`cortisone itself has some palliative and anti-tumor
`activity [28–30]. In conclusion, the lessons learned
`from the development of suramin can be applied to
`all clinical trials in hormone refractory prostate can-
`cer. Suramin has significantly enhanced the evolution
`of our knowledge in several areas of prostate cancer
`biology and treatment.
`
`Acknowledgements
`
`We would like to thank Catherine Morris for assistance
`in the preparation of this manuscript.
`This work has been funded by the US Government,
`the NIH’s Office of Research on Minority Health.
`
`References
`
`1. Armand JP, Cvitkovic E: Suramin: a new therapeutic concept.
`Eur J Cancer 26: 417–419, 1990
`2. Walther MM, Figg WD, Lineham WM: Intravesical suramin:
`a novel agent for the treatment of superficial transitional-cell
`carcinoma of the bladder. World J Urol 14 Suppl 1: S8–11,
`1996
`3. Eisenberger MA, Sinibaldi V, Reyno L: Suramin. Cancer Pract
`3: 187–189, 1995
`4. Broder S, Yarchoan R, Collins JM, Lane HC, Markham PD,
`Klecker RW, Redfield RR, Mitsuya H, Hoth DF, Gelmann E:
`Effects of suramin on HTLV-III/1AV infection presenting as
`Kaposi’s sarcoma or AIDS-related complex: clinical pharma-
`cology and suppression of virus replication in vivo. Lancet 2:
`627–630, 1985
`
`

`
`218
`
`5. Cheson BD, Levine AM, Mildvan D, Kaplan LD, Wolfe P,
`Rios A, Groopman JE, Gill P, Volberding PA, Poiesz BJ:
`Suramin therapy in AIDS and related disorders. JAMA 258:
`1347–1351, 1987
`6. Armand JP: New anticancer drugs in Europe. Chin Med J 110:
`297–308, 1997
`7. Eisenberger MA, Fontana JA: Suramin, an active non-
`hormonal cytotoxic drug for treatment of prostate cancer:
`compelling reasons for testing in patients with hormone re-
`fractory breast cancer. J Natl Cancer Inst 84: 3–5, 1992
`8. Myers C, Cooper M, Stein C, LaRocca R, Walther MM,
`Weiss G, Choyke P, Dawson N, Steinberg S, Uhrich M:
`Suramin: a novel growth factor antagonist with activity in
`hormone-refractory metastatic prostate cancer. J Clin Oncol
`10: 881–889, 1992
`9. Stein CA, LaRocca R, Thomas R, McAtee N, Horne M, Myers
`CE: Suramin – an anticancer drug with a unique mechanism
`of action. J Clin Oncol 7: 449–508, 1989
`10. Myers C, LaRocca R, Cooper M: Role of Suramin in Can-
`cer Biology and Treatment. Baltimore, Williams and Wilkins,
`1991.
`11. Scher HI, Steineck G, Kelly WK: Hormone-refractory (D3)
`prostate cancer: refining the concept. Urology 46: 142–148,
`1995
`12. Eisenberger MA, Reyno LM, Jodrell DI, Sinibaldi VJ,
`Tkaczuk KH, Sridhara R, Zuhowski EG, Lowitt MH, Jacobs
`SC, Egorin MJ: Suramin, an active drug for prostate cancer:
`interim observations in a phase I trial. J Natl Cancer Inst 85:
`611–621, 1993
`13. Kobayshi K, Vokes EE, Vogelzang NJ, Janisch L, Soliven B,
`Ratain MJ: Phase I study of suramin given by intermittent infu-
`sion without adaptive control in patients with advanced cancer.
`J Clin Oncol 13: 2196–2207, 1995
`14. Panvichian R, Pienta KJ: The treatment of hormone refractory
`prostate cancer. Compr Ther 22: 81–87, 1996
`15. Small EJ, Meyer M, Marshall ME, Reyno L, Meyers F, Natale
`R, Meyer M, Lenehan P, Chen L, Slichenmyer WJ, Eisen-
`berger M: Suramin therapy for patients with symptomatic
`hormone refractory prostate cancer; results of a phase III
`trial comparing suramin plus hydrocortisone to placebo plus
`hydrocortsone. J Clin Oncol 18 (7): 1440–1450, 2000
`16. Bubley GJ, Carducci M, Dahut W, Dawson N, Daliani D, Eis-
`enberger M, Figg WD, Freidlin B, Halabi S, Hudes G, Hussain
`M, Kaplan R, Myers C, Oh W, Petrylak DP, Reed E, Roth B,
`Sartor O, Scher H, Simons J, Sinibaldi V, Small EJ, Smith MR,
`Trump DL, Vollmer R, Wilding G: Eligibility and response
`guidelines for phase II clinical trials in androgen independent
`prostate cancer: Recommendations from the Prostate-Specific
`Antigen Working Group. J Clin Oncol 17: 3461–3467, 1999
`17. Thalmann GN, Sikes RA, Chang S, Johnston DA, VonEschen-
`bach AC, Chung WK: Suramin-induced decrease in prostate-
`specific antigen expression with no effect on tumor growth in
`the LNCaP model of human prostate cancer. J Natl Cancer Inst
`88: 174–801, 1996
`18. Cooper EH, Robinson MR, Whelan P, Ferro MA: Tumor
`markers in prostate cancer. Cancer 70 (Suppl 1): 225–229,
`1992
`19. Ferro MA, Barnes I, Roberts JB, Smith PJ: Tumour marker