Tanespimycin monotherapy in relapsed multiple myeloma: results of a phase 1 dose-escalation study

Summary

Tanespimycin, a heat shock protein 90 (HSP90) inhibitor, induces apoptosis in drug-sensitive and -resistant MM cell lines and in tumour cells from patients with relapsed MM. In this phase 1 dose-escalation study, the safety, plasma pharmacokinetics, and biological/antitumour activity of tanespimycin were evaluated in heavily pretreated patients with relapsed/refractory MM. Tanespimycin (150–525 mg/m²) was given on days 1, 4, 8, and 11 of each 3-week cycle for up to 8 cycles. Non-haematological AEs included diarrhoea (59%), back pain (35%), fatigue (38%), and nausea (35%); haematological AEs included anaemia (24%) and thrombocytopenia (21%). One patient (3%) achieved minimal response (MR), with a progression-free survival (PFS) of 3 months, a 41% decrease from baseline in urine M protein, and a 33% decrease from baseline in serum M protein. Fifteen patients (52%) achieved SD with a median PFS of 2·1 months; 5/15 had reductions in serum M protein ranging from 7% to 38% and in urine M protein ranging from 6% to 91%. Mean HSP70 levels increased from day 1 h 0 to day 1 h 4 with further increases on day 11 h 0 and day 11 h 4, consistent with a therapeutic treatment effect. Tanespimycin monotherapy was well tolerated and demonstrated activity across all doses tested.

Heat shock protein 90 (HSP90), one of a family of molecular chaperones, is intimately involved in the survival of tumour cells. Under normal conditions, HSP90 prevents aggregation of proteins and functions as a chaperone that preserves the 3-dimensional conformation, intracellular localization, and activity and regulates the proteolytic turnover of a range of proteins. Under conditions of stress, the expression of HSP90 increases as an adaptive response intended to enhance cell survival. This basic function also contributes to tumour cell proliferation, thereby driving oncogenesis. (Whitesell & Lindquist, 2005; Powers & Workman, 2006).

HSP90 is uniformly expressed in a variety of human multiple myeloma (MM) cell lines and primary specimens from patients with MM. (Mitsiades et al, 2002, 2006) It has been shown to be critical to the survival of MM cell lines grown in culture. (Chatterjee et al, 2007) Investigators have also demonstrated that HSP90 inhibitors suppress the expression of client proteins including insulin-like growth factor-1 receptor (IGF-1R) and interleukin 6 receptor (IL-6R), which in turn disrupts interactions between bone marrow stromal cells and MM tumour cells,(Mitsiades et al, 2006) as well as upregulating other HSPs including HSP70. Upregulation of HSP70 is a marker of HSP90 inhibition, and multiple phase 1 clinical trials evaluating HSP90 inhibitors have shown a correlative increase in HSP70 in the presence of HSP90 inhibition. (Banerji et al, 2005; Goetz et al, 2005; Grem et al, 2005).

Tanespimycin (17-allylamino-17-demethoxygeldanamycin, 17-AAG) is a synthetic geldanamycin analogue and was the first HSP90 inhibitor to enter clinical trials(Sausville et al, 2003) in patients with advanced malignancies. (Banerji et al, 2005; Goetz et al, 2005; Grem et al, 2005; Ramanathan et al, 2007) Studies have evaluated several dosing schedules in this patient population to identify a maximum tolerated dose (MTD). (Banerji et al, 2005; Goetz et al, 2005; Grem et al, 2005; Nowakowski et al, 2006; Solit et al, 2007) One trial infused tanespimycin on days 1, 8, and 15 of a 28-day cycle(Goetz et al, 2005); another chose a daily dosing schedule of 5 d every 3 weeks(Grem et al, 2005); and a third used a once weekly regimen. (Banerji et al, 2005) These dosing schedules were generally well tolerated and produced antitumour activity in various solid tumour types.

As part of in vitro studies, tanespimycin was shown to potently induce apoptosis of both drug-sensitive and drug-resistant MM cell lines, as well as tumour cells from patients with relapsed/refractory MM. (Mitsiades et al, 2006) Tanespimycin was also shown to suppress cell surface expression of receptors for cytokines mediating MM cell growth, survival, drug resistance, and related downstream signalling pathways among other pleiotropic antiproliferative and proapoptotic molecular sequelae, allowing tanespimycin to sensitize MM cells to other anticancer agents. (Mitsiades et al, 2006).

Based on the preclinical rationale and safety profile demonstrated in phase 1 trials, the primary study objective was to evaluate the MTD of tanespimycin given in a twice weekly dosing schedule (day 1, 4, 8, and 11 of each 21-day cycle), for up to 8 cycles of treatment in patients with relapsed and refractory MM. Secondary objectives were to measure plasma pharmacokinetics (PK) of tanespimycin in the study population, evaluate the safety of repeated tanespimycin doses, and assess any preliminary evidence of antitumour activity following treatment. Results showed both anti-disease activity and favorable tolerability, establishing the framework for subsequent clinical exploration as monotherapy and as part of combination regimens.

Methods

Patients

Patients were considered eligible if they had relapsed or relapsed and refractory MM following ≥ 2 prior regimens (including stem cell transplant), were at least 18 years old, with a Karnofsky performance status (KPS) of ≥ 70%, and histological evidence of disease at diagnosis. Relapse from prior therapy was determined by the modified European Group for Blood and Marrow Transplantation (EBMT) criteria (Bladé et al, 1998, Richardson et al, 2003), including reappearance of serum or urinary paraprotein. Relapsed and refractory MM was defined as progression or nonresponse during salvage chemotherapy or within 60 days after completion of treatment. (Richardson et al, 2003; Anderson, et al 2008) Patients were excluded if they had pre-existing severe neuropathy (Grade ≥ 3 as defined by the National Cancer Institute Common Terminology Criteria for Adverse Events [NCI CTCAE] v3.0 [http://ctep.cancer.gov/protocolDevelopment/electronic_applications/docs/ctcaev3.pdf]). Patients with central nervous system involvement from MM, concomitant medications that could prolong QTc interval, prior radiation involving the heart, or history of significant heart disease including, but not limited to, myocardial infarction, left bundle branch block, and congenital history of QTc prolongation were excluded from the study.

Study design and treatment

This was a phase 1, open-label, dose-escalation trial in patients with relapsed MM. The objectives of the study were to evaluate the safety, plasma PK, and biological and antitumour activity of tanespimycin.

Tanespimycin was administered intravenously over 60–120 minutes. Patients received infusions of the standard Cremophor-based formulation of tanespimycin on days 1, 4, 8, and 11 in each 3-week cycle and were treated for up to 8 cycles until disease progression or unacceptable toxicity. The doses tested include 150, 220, 275, 340, 420, and 525 mg/m².

Three patients were assigned to each cohort. If no dose-limiting toxicity (DLT) was observed in the three evaluable patients of the cohort, the study proceeded to the next dose level. If 1 of 3 evaluable patients experienced a DLT, then the cohort was increased to 6 evaluable patients. If no more than 1 of the 6 patients experienced a DLT, then the next dose level was evaluated. During cycle 1, DLTs were defined as: Grade 4 neutropenia, anaemia, or thrombocytopenia; any Grade 3 or greater non-haematological toxicity (except Grade 3 injection site reaction, alopecia, anorexia, fatigue, or Grade 3 nausea, diarrhoea, or vomiting that did not receive maximal supportive care); treatment delay of more than 4 weeks due to prolonged recovery from a drug-related toxicity; and dose modification based on newly observed cardiac toxicity.

Pharmacokinetics

PK sampling was done during the first treatment cycle. Blood samples were collected on day 1 and day 11 to determine plasma concentrations of tanespimycin and its major metabolite, 17-AG. Blood was drawn before treatment, 30 min after the start of infusion, immediately prior to the end of infusion, and 0·08, 0·25, 0·5, 1, 2, 4, 8, and 24 h after the end of infusion.

PKs were calculated using compartmental-independent analysis with the Kinetica^® software package (Thermo Fisher Scientific Inc., Waltham, MA, USA). The following PK parameters were calculated using standard noncompartmental analysis: maximum plasma concentration (Cmax), time to maximum plasma concentration (Tmax), area under the curve (AUC), half-life (t1/2), and total systemic clearance (CL). Twenty-nine patients had blood collected for PKs on day 1 and 23 patients on day 11.

Efficacy and safety analyses

Although an assessment of efficacy was not the primary study objective, tumour response was determined based on the modified EBMT/International Bone Marrow Transplant Registry (IBMTR) criteria, including serum M protein (M protein spike) and urine M protein values, bone marrow aspirate, and extramedullary tumour imaging, and on progression-free survival (PFS). (Bladé et al, 1998; Richardson et al, 2003) Investigators monitored antitumour responses for complete response (CR), near complete response (nCR), partial response (PR), minimal response (MR), stable disease (SD), and progressive disease (PD). MR was determined by at least 25% reductions in serum M protein, urine M protein, soft tissue plasmacytomas, and absence of increase in lytic bone lesions. (Anderson, et al 2008) SD was determined when the patient did not meet the criteria for CR, PR, MR, or PD. Reductions in monoclonal paraprotein were also evaluated. All patients who received at least 1 dose of tanespimycin were included in the safety analysis. The safety data include frequency of adverse events (AEs) and frequency and shift of laboratory variables, including neutrophil count, alanine aminotransferase (ALT), and aspartate aminotransferase (AST).

Laboratory determinations were performed prior to first and last drug administration of each cycle, and included complete blood count with differential and platelet counts, clinical chemistry, coagulation tests, and urinalysis. Potassium, calcium, and magnesium were analysed 24 hours before each infusion and verified to be in their respective normal ranges. Electrocardiograms (ECGs) were performed at baseline and at the end of the study. Additional ECGs were performed at the discretion of the investigator as clinically indicated.

Biomarker analysis

Peripheral blood mononuclear cells (PBMCs) were obtained prior to treatment initiation and at 4 h post dosing on day 1 and day 11 and examined for change in HSP70 expression using Western blot analysis.

Results

Patient demographics and baseline clinical characteristics for the 29 patients enrolled in the study are shown in Table I. The median age of the patients was 63 years, and 22 (76%) were men. The median time since diagnosis was 41 months (range, 11–152 months) and most patients (62%) had a KPS of 90%. The most frequent heavy chain subtype was IgG, observed in 22 patients (76%) and median β₂microglobulin was 3·7 mg/l (range, 1–15).

Table I.

Demographics and baseline clinical characteristics.

Category	Tanespimycin dose, mg/m2
	150 n = 4	220 n = 9	275 n = 3	340 n = 7	420 n = 3	525 n = 3	All dose cohorts N = 29
Age (years), median	58	63	54	71	58	63	63
Range	50–64	46–79	43–66	58–80	48–58	47–75	43–80
Sex
Male, n (%)	4 (100)	6 (67)	2 (67)	5 (71)	3 (100)	2 (67)	22 (76)
Ethnicity, n (%)
White	3 (75)	8 (89)	3 (100)	7 (100)	3 (100)	2 (67)	26 (90)
Black	1 (25)	0	0	0	0	1 (33)	2 (7)
Hispanic	0	1 (11)	0	0	0	0	1 (3)
KPS, median	90	90	90	90	90	90	90
Range	80–90	80–100	90–90	70–90	80–90	85–100	70–100
Subtype, n (%)
IgG	3 (75)	8 (89)	2 (67)	4 (57)	3 (100)	2 (67)	22 (76)
IgA	1 (25)	1 (11)	1 (33)	3 (43)	0	0	6 (21)
Nonsecretory	0	2 (22)	1 (33)	1 (14)	0	0	4 (14)
β2 microglobulin, median (mg/l)	5·8	6·2	2·4	3·8	2·4	3·2	3·7
Median months since diagnosis	39	41	26	48	20	86	41
Any prior stem cell transplant, n (%)	3 (75)	3 (33)	0	3 (43)	1 (33)	2 (67)	12 (41)
Median number of prior regimens	4	5	4	4	3	6	4
Range	3–6	2–19	4–5	3–8	3–4	3–7	2–19
Prior bortezomib, n (%)	4 (100)	8 (89)	2 (67)	5 (71)	2 (67)	2 (67)	23 (79)
Prior lenalidomide, n (%)	2 (50)	3 (33)	1 (33)	2 (29)	0	0	8 (28)
Prior thalidomide, n (%)	2 (50)	8 (89)	3 (100)	6 (86)	3 (100)	3 (100)	25 (86)

KPS, Karnofsky performance status.

All 29 of the study participants were relapsed and refractory, as defined by the enrollment criteria. Thirteen patients (49%) had undergone prior stem cell transplant and patients had received a median of 4 prior treatment regimens (range, 2–19), which included thalidomide (86%), bortezomib (79%), vincristine and doxorubicin as part of VAD (vincristine, doxorubicin, and dexamethasone; 60%), dexamethasone (41%), and lenalidomide (28%). Of the 29 evaluable patients, 14 had progressed on their most recent prior bortezomib-based or lenalidomide-based therapy (11 from bortezomib-based therapy, 4 from lenalidomide-based therapy, and 1 who had been treated with both drugs and progressed on each). The remaining participants had relapsed and proven refractory on the other chemotherapy regimens listed above.

There were 4 patients in the 150 mg/m², 9 in the 220 mg/m², 3 in the 275 mg/m², 7 in the 340 mg/m², and 3 each in the420 mg/m² and 525 mg/m² cohort, respectively. The entire population (N = 29) completed a median of 3 (range, 1–8) cycles of treatment.

Pharmacokinetics

PK parameters showed that median Cmax increased with increasing doses and median Tmax ranged from 1 to 2 h (Table II). The t1/2 ranged from 1·18 (220 mg/m²) to 5·16 h (275 mg/m²). The mean (% coefficient of variation [CV]) tanespimycin concentrations at 24 h after dose 1 was 2·7 (200), 2·8 (199), 3·4 (173), 8·5 (141), 9·9 (89), and 24·2 ng/ml (74) for the 150, 220, 275, 340, 420, and 525 mg/m² cohorts, respectively. The mean (% CV) 17-AG metabolite concentrations at 24 h after dose 1 were 46·4 (110), 27·5 (64), 54·2 (21), 65·7 (34), 61·4 (79), and 93·1 ng/ml (99) for the 150, 220, 275, 340, 420, and 525 mg/m² cohorts respectively.

Table II.

Pharmacokinetic Parameters of Tanespimycin^*.

Dose, mg/m2	Statistic	Cmax (ng/ml)	Tmax (h)	AUCtot (ng/ml* h)	t1/2 (h)	CL (l/h)
150 (n = 4)	Mean	2403	0·99	5592·3	2·19	54·92
	Median (min–max)	2435 (1470–3220)	0·99 (0·87–1·08)	5517·3 (4720·8–7020·8)	2·06 (1·45–3·41)	56·40 (43·44–63·15)
220 (n = 9)	Mean	4590	1·03	9405·0	2·68	46·77
	Median (min–max)	4050 (3080–8540)	1·00 (0·98–1·25)	9505·9 (4521·0–15,848·8)	4·40 (1·18–2·36)	74·98 (19·31–47·90)
275 (n = 3)	Mean	6820	1·03	13,802·8	2·96	39·63
	Median (min–max)	7065 (4720–8530)	1·00 (0·98–1·15)	14,250·0 (11,867·5–15,708·5)	2·91 (1·58–5·16)	39·74 (28·39–49·80)
340 (n = 7)	Mean	8751	0·99	17,936·1	3·10	37·20
	Median (min–max)	7600 (6740–14,600)	0·98 (0·50–1·33)	15,615·1 (10,380·6–32,100·5)	3·30 (1·80–4·30)	36·80 (22·70–54·00)
420 (n = 3)	Mean	7133	1·58	23,069·7	3·00	41·50
	Median (min–max)	8500 (4470–7570)	2·28 (1·00–1·48)	36,003·7 (13,124·4–24,482·8)	4·08 (2·30–2·93)	69·41 (22·05–34·87)
525 (n = 3)	Mean	9920	2·00	29,639·0	2·60	57·0
	Median (min–max)	10,600 (5680–12,800)	2·20 (2·00–2·80)	28,683·0 (17,375·0–43,816·0)	2·60 (2·20–3·00)	57·0 (35·6–78·3)

AUCtot, area under the concentration time curve from time zero to infinity; CL, total systemic clearance; Cmax, maximum plasma concentration; t1/2, half-life; Tmax, time to maximum plasma concentration.

^*Distributive volume values and clearance reported have been corrected for body surface area.

The plasma concentration over time profiles across doses are shown in Fig 1. Tanespimycin displayed linear kinetics over the dose range 150–525 mg/m². Plasma concentration over time profiles decayed in a bi-exponential manner with a rapid distribution phase and slower elimination phase. The major metabolite of tanespimycin, 17-AG, displayed a prolonged plasma concentration over time profile compared with the parent drug, with significant levels detected at 24 h (data not shown). Comparison of the PK parameters in individual patients on days 1 and 11 showed no accumulation of drug in plasma. The PK of tanespimycin was stable at dosing intervals of 4 d with no apparent induction of metabolism.

Fig 1.

Plasma concentration over time profiles for tanespimycin were generated from blood samples drawn during cycle 1.

Safety

AEs occurring in ≥20% of all patients and selected AEs of interest are shown in Table III. The most frequent non-haematological AEs were diarrhoea (59%), back pain (35%), fatigue (38%), and nausea (35%). There were 7 cases of increased ALT (two of Grade 3), and 9 cases of increased AST (two Grade 3; one Grade 4); the Grade 3/4 AEs were also DLTs. Twenty-three patients had a history of peripheral neuropathy and 1 patient (1/23) in the 340 mg/m² reported treatment-emergent Grade 2 neuropathy; there were no other reports of peripheral neuropathy during therapy.

Table III.

AEs Occurring in ≥20% of all patients and selected AEs of interest.

AEs, n (%)	Tanespimycin dose, mg/m2
	150 n = 4	220 n = 9	275 n = 3	340 n = 7	420 n = 3	525 n = 3	All dose cohorts N = 29
Any	4 (100)	9 (100)	3 (100)	7 (100)	3 (100)	3 (100)	29 (100)
Non-haematological*
Diarrhoea	2 (50)	3 (33)	3 (100)	4 (57)	3 (100)	2 (67)	17 (59)
Back pain	3 (75)	1 (11)	2 (67)	2 (29)	2 (67)	0	10 (35)
Fatigue	1 (25)	2 (22)	2 (67)	2 (29)	2 (67)	2 (67)	11 (38)
Nausea	1 (25)	1 (11)	1 (33)	2 (29)	3 (100)	2 (67)	10 (35)
AST increased	1 (25)	1 (11)	1 (33)	2 (29)	2 (67)	2 (67)	9 (31)
Muscle spasms	1 (25)	3 (33)	0	3 (43)	1 (33)	0	8 (28)
ALT increased	1 (25)	1 (11)	0	2 (29)	2 (67)	1 (33)	7 (24)
Constipation	2 (50)	1 (11)	0	2 (29)	1 (33)	0	6 (21)
Dyspnea	0	1 (11)	0	4 (57)	0	1 (33)	6 (21)
Pain in extremity	2 (50)	3 (33)	0	0	1 (33)	0	6 (21)
Neuropathy	0	0	0	1 (14)	0	0	0
Haematological
Anaemia	1 (25)	3 (33)	1 (33)	1 (14)	1 (33)	0	7 (24)
Thrombocytopenia	1 (25)	1 (11)	0	3 (43)	1 (33)	0	6 (21)
Neutropenia	0	0	0	0	0	0	0

AEs, adverse events; ALT, alanine aminotransferase; AST, aspartate aminotransferase.

^*AEs affecting at least 20% of the total group.

The most frequent haematological AEs were anaemia (24%) and thrombocytopenia (21%). There were no instances of neutropenia observed. The most common Grade 3/4 AEs were 3 cases of anaemia (Grade 3) and 2 cases of thrombocytopenia (Grade 4). Five of 29 patients experienced DLTs: 1 patient each in the 150 and 340 mg/m² groups with Grade 4 thrombocytopenia,1 patient in the 220 mg/m² group with Grade 3 ALT, 1 patient in the 340 mg/m² group with Grade 4 elevated AST and Grade 3 elevated ALT, and 1 patient in the 525 mg/m² group with Grade 3 cardiac ischaemia and Grade 2 abnormal coordination, although this was considered unrelated to study drug by the treating physician.

In cycles 1–8, there was no meaningful change from baseline in mean absolute neutrophil count (all cycles, ≤0·079 × 10⁹/l) and no meaningful change in mean haemoglobin levels was observed during multiple treatment cycles.

In the first 4 cycles, mean platelet counts showed a decrease between 10 × 10⁹ and 20 × 10⁹/l from baseline to day 11 and then returned to baseline levels by day 1 of the subsequent cycle. Subsequent cycles demonstrated a similar pattern. No QTc prolongation or changes in left ventricular ejection fraction were observed. No patients had QTc increases of 60 ms or more (as amended using a QT interval per the Fridericia correction formula).

Efficacy

Responses observed during the study according to modified EBMT criteria (Bladé et al, 1998; Richardson et al, 2003; Anderson, et al 2008) and M protein reduction are shown in Table IV. None of the patients had CR, nCR, or PR. One patient (3%) in the 150 mg/m² dose group achieved a MR. An additional 15 patients (52%) who met neither the criteria for therapeutic response nor exhibited PD were determined to have SD. Eleven of the remaining 13 patients (38%) had PD, and 2 patients withdrew prior to treatment. The patient with the durable MR received four prior chemotherapies including a best response of PD to lenalidomide (1 month prior to study start) preceeded by a best response of SD to prior bortezomib/dexamethasone (8 months prior to study start). The MR patient had a 41% decrease in urine M protein level and a 33% decrease from baseline in serum M protein level. Five SD patients had reductions in serum M protein ranging from 7% to 38% and reductions in urine M protein ranging from 6% to 91%. The MR patient had a PFS of 3 months. The 15 SD patients had a median PFS of 2·1 months (95% confidence interval, 1·48–2·60). The median PFS for the 13 non-responding patients was 1·4 months (range, 0·5–3·0).

Table IV.

Responses by EBMT Criteria and Other Evidence of Activity.

Category	Tanespimycin dose, mg/m2
	150 n = 4	220 n = 9	275 n = 3	340 n = 7	420 n = 3	525 n = 3	All dose cohorts N = 29
Minimal response, n (%)	1 (25)	0	0	0	0	0	1 (3)
Stable disease, n (%)	2 (50)	5 (56)	3 (100)	3 (43)	1 (33)	1 (33)	15 (52)
Progression of disease, n (%)	1 (25)	4 (44)	0	3 (43)	2 (67)	1 (33)	11 (38)
Missing, n (%)	0	0	0	1 (14)	0	1 (33)	2 (7)
Decrease in M protein, n	1	0	1	2	0	0	4
Decrease in M spike, %	33		11	20, 11
Decrease in urinary M protein, %	41		12	91, 43

EBMT, the European Group for Blood and Marrow Transplantation.

Biomarker analysis

The mean HSP70 level, a biomarker of HSP90 inhibition, increased in the 275–420 mg/m² cohorts from day 1 h 0, to day 1 h 4 with further increases observed on day 11 h 0 and day 11 h 4 (Fig 2). The increase was most marked in the 275 mg/m² cohort but evidence of upregulation of HSP70 was also present in the cohorts.

Fig 2.

Changes in HSP70 expression in peripheral blood monocytes using Western blot analysis. Western blot analysis was performed on lysates from peripheral blood mononuclear cells obtained on day 1 and day 11 of the first cycle.

Discussion

In this phase 1 study, tanespimycin showed evidence of activity as a monotherapy in heavily pretreated patients with relapsed and refractory MM. The most common AEs included diarrhoea, back pain, fatigue, and nausea. The frequency of elevated transaminases (a reflection of hepatic toxicity) was low and caused only 1 patient to discontinue the study. Anaemia and thrombocytopenia occurred in approximately 20% of patients, but the frequency of neutropenia was remarkably low. Thrombocytopenia seemed to be transient with platelet counts returning to baseline between days 11 and 21. Importantly, no severe peripheral neuropathy was observed. In addition, the frequency of Grade 3/4 AEs was low and only 3 patients discontinued due to AEs. Tanespimycin was thus generally well tolerated in this patient population.

This study enrolled heavily pretreated patients who were relapsed and refractory after at least two prior therapies. Despite their advanced and resistant disease status, anti-tumour activity was evidenced by SD in 52% of patients with durable MR in 1 patient (3%), despite patients having progressed on bortezomib- and lenalidomide-based therapies.

Activity was seen in patients across all tanespimycin dosage ranges, with stabilization of disease occurring in 11/16 patients (63%) who received the three lowest doses (150, 220, and 275 mg/m²) and in 5/13 patients (38%) who received the three highest doses (340, 420, and 535 mg/m²). These results provide preliminary evidence that tanespimycin doses as low as 150 mg/m² have activity in at least some patients, and that a less intensive schedule of administration may be efficacious while minimizing the potential for adverse effects. Results showing a minimal therapeutic response (in the MR patient) or activity without a measurable paraprotein reduction (in the SD patients) may be considered modest overall. However, as noted by other investigators, MM treatment outcomes below the threshold for regulatory approval may still be an indicator of drug activity, especially in patients who were poorly responsive to prior therapy. (Anderson, et al 2008) Specifically, the failure of prior therapeutic regimens as an indication of rapidly progressing disease suggest that these patients would be poorly responsive to other treatments. To observe MR or better with tanespimycin monotherapy is thus noteworthy.

Taken together, the AEs, haematological, and clinical chemistry data (Table III) did not establish a MTD or determine a DLT for tanespimycin in this trial. Various AEs of relatively minor severity occurred across all dose ranges, suggesting a more idiosyncratic pattern. Importantly, the antineoplastic activity seen at the three lowest doses provides evidence that a therapeutic response can be obtained at lower doses that carry minimal toxicity risk. This evidence in part decreases the need to evaluate relatively higher and potentially more toxic tanespimycin doses as monotherapy, and provides a useful platform for combination approaches.

Biological activity of tanespimycin was also observed as reflected by increases in HSP70, a biomarker for HSP90 inhibition, and seen at all tanespimycin doses on day 1 and day 11 (Fig 2). The increased levels of HSP90 on day 11, 3 d since the previous dose, support the dosing regimen intervals. These results are similar to a prior study showing HSP70 induction at 6 h following a dose of at least 80 mg/m² per week. (Banerji et al, 2005) Similar findings have been noted in other phase 1 trials. (Goetz et al, 2005; Grem et al, 2005) Specifically, Modi et al (2007) showed that increases in HSP70 observed during treatment with tanespimycin demonstrate that this agent effectively targets HSP90 at doses as low as 275 mg/m² and that intermittent dosing is appropriate based on the continued increase of HSP70 to day 11 relative to day 1.

The patient population evaluated in this trial was extensively pretreated and all had relapsed and refractory MM. By comparison, in prior phase 2 studies of bortezomib monotherapy in relapsed or refractory MM, the median number of prior regimens was 3–6, with a range of 1–7. (Richardson et al, 2003; Jagannath et al, 2004) In a phase 3 study of bortezomib monotherapy versus high-dose dexamethasone in less advanced MM, the median number of prior regimens was 2, (Richardson et al, 2005) and in a phase 3 study of pegylated liposomal doxorubicin plus bortezomib, 34% had only 1 prior therapy. (Orlowski et al, 2007) In studies with lenalidomide, patient populations have been similar; in a phase 3 study of lenalidomide plus dexamethasone in relapsed and refractory MM, approximately 30% of patients in each arm had only 1 prior therapy. (Dimopoulos, et al 2007).

Patients with relapsed and refractory MM, in whom several lines of prior therapy have failed, are inherently difficult to treat and represent a patient population with a considerable unmet need for new and effective treatment options. Currently approved novel agents, such as bortezomib and lenalidomide, trigger a state of stress in MM cells, presumably making them more dependent on HSP90 for survival. Therefore, inhibiting HSP90 in the presence of bortezomib, for example, prevents the MM cell from compensating for bortezomib-induced stress, thus favoring apoptosis. This hypothesis is supported by preclinical evidence that the combination of tanespimycin and bortezomib is more effective than either agent alone and is the rationale for the use of these two drugs clinically in MM (Mitsiades et al, 2006).

Tanespimycin in combination with bortezomib has since been successfully evaluated in patients with relapsed or relapsed and refractory MM as part of a phase 1/2 study. (Richardson et al, 2006, 2009).

Conclusion

Tanespimycin was generally well tolerated and demonstrated therapeutic plasma concentrations as well as target inhibition. The findings in this phase 1 study are in alignment with other phase 1 studies of tanespimycin and support ongoing trials in patients with MM in combination with other agents including bortezomib, as well as other advanced malignancies.