• Open Access

Double-Blind Placebo-Controlled Trial of Adjuvant Pamidronate with Palliative Radiotherapy and Intravenous Doxorubicin for Canine Appendicular Osteosarcoma Bone Pain

Authors


  • This study was conducted at the Veterinary Teaching Hospital, University of Illinois, Urbana, IL.

Corresponding author: Timothy M. Fan, DVM, PhD, Department of Veterinary Clinical Medicine, University of Illinois at Urbana-Champaign, 1008 West Hazelwood Drive, Urbana, IL 61802-4714; e-mail: t-fan@uiuc.edu.

Abstract

Background: Canine osteosarcoma (OSA) causes focal malignant osteolysis leading to severe pain. Despite the documented efficacy of radiotherapy or IV aminobisphosphonates for managing cancer bone pain, their potential combined therapeutic value has not been reported in OSA-bearing dogs.

Hypothesis: Pamidronate combined with standardized palliative therapy will improve pain control and bone biologic effects in OSA-bearing dogs.

Animals: Fifty dogs with appendicular OSA treated with standardized palliative therapy and either pamidronate or sterile saline.

Methods: Randomized, prospective, double-blinded, placebo-controlled study. Treatment responses for dogs receiving standardized palliative therapy with (n = 26) or without (n = 24) adjuvant pamidronate were serially evaluated for changes in subjective pain scores, urine N-telopeptide (NTx) excretion, primary tumor relative bone mineral density (rBMD), and computerized pressure platform gait analysis.

Results: Median duration of subjective pain relief for dogs treated with adjuvant pamidronate or placebo was 76 and 75 days, respectively (P= .39). Forty percent (20/50; pamidronate [11/26] and placebo [9/24]) of dogs experienced durable analgesia, defined by pain alleviation ≥112 days. For patients achieving durable pain control, dogs treated with pamidronate achieved greater reductions in NTx excretion and larger increases in rBMD compared with placebo controls. Changes in peak vertical force assessed by computerized pressure platform gait analysis correlated with pain alleviation in OSA-bearing dogs.

Conclusions and Clinical Importance: Combining pamidronate with standardized palliative therapy is safe, but does not clearly improve pain alleviation. However, in dogs achieving durable pain control, adjuvant pamidronate appears to decrease focal bone resorption in the local tumor microenvironment.

In a palliative setting, achievement of durable pain alleviation is the limiting factor for dogs with appendicular osteosarcoma (OSA). Pain associated with skeletal neoplasms is attributed to 2 specific host responses. First, the presence of cancer cells in the bone microenvironment results in the release of chemical mediators by nonneoplastic stromal cells which in turn stimulate periosteal nociceptors, leading to the generation of painful sensations.1,2 Second, the genesis, maintenance, and exacerbation of bone cancer pain may be directly attributed to dysregulated and pathologic osteoclastic bone resorption.3–5 Based upon these mechanisms of bone pain generation, the optimal management of malignant osteolytic pain would combine eradication of cancer cells in bone matrix with inhibition of tumor-induced osteoclastic bone resorption.

Conventional therapy directly targeting tumor cells in bone matrix consists primarily of ionizing radiation therapy.6 In veterinary oncology, different sources of ionizing radiation have been reported to be effective in palliating malignant osteolysis associated with primary or metastatic lesions.7–12 In dogs with OSA, the analgesic effects of ionizing radiation likely are attributed to apoptosis of tumor cells and normal bone stromal cells, which are jointly responsible for stimulating periosteal nociceptors in the tumor microenvironment.13,14 In addition, some clinical evidence suggests that dogs with OSA may derive more durable pain relief when systemic chemotherapy is coadministered with radiotherapy.8,12

Aminobisphosphonates interfere with the geranylgeranylation of small GTPase proteins involved in cell signaling, thereby impeding osteoclast activity and inducing osteoclast apoptosis.15,16 As a result, IV aminobisphosphonates are 1st-line agents in the treatment of malignant osteolytic pain in people.17 Similar to their analgesic and antiresorptive effects in people with skeletal malignancies, single-agent IV aminobisphosphonates have shown beneficial analgesic and bone biologic effects in dogs with appendicular OSA and other skeletal malignancies.18–20

For dogs with malignant bone pain, analgesia may be assessed by serial evaluation of subjective and objective endpoints. Subjective measures of analgesic response include visual analog scale (VAS) or cumulative pain index score (CPIS) assessments.18,20,21 Although easy to implement in a clinical setting, such subjective measures often are associated with high inter- and intraobserver variability because of pet owner and clinician bias. Because the severity of bone cancer pain is highly correlated with the extent of malignant osteolysis,5 objective measures of therapeutic response potentially include quantification of bone resorption markers such as urine N-telopeptide (NTx), which reflects dynamic fluctuations in global skeletal metabolism.22,23 Additionally, methods quantifying relative bone mineral density (rBMD) of focal osteolytic lesions after therapy, such as dual energy X-ray absorptiometry (DEXA), can provide biologically relevant information regarding skeletal changes in the local tumor microenvironment.24–26 Finally, because dogs with appendicular OSA primarily are presented for lameness secondary to pain, improvements or normalization of ambulation assessed by computerized pressure platform gait analysis may serve as a reliable and objective clinical measure of pain palliation.27

For OSA-bearing dogs receiving palliative therapy, achievement of effective and durable analgesia remains of utmost importance to owners seeking to improve or maintain their pets' quality of life. Given the documented palliative effectiveness of ionizing radiation and IV aminobisphosphonates as monotherapy for dogs with appendicular OSA, combining these 2 treatment modalities appears clinically justified. The purpose of this study was to investigate if the addition of adjuvant pamidronate to a standardized palliative protocol would improve clinical pain alleviation, enhance beneficial bone biologic activity, and improve gait in dogs with appendicular OSA.

Materials and Methods

Study Population

Fifty dogs were prospectively evaluated between September 2004 and March 2008 at the University of Illinois, Cancer Care Clinic (UICCC). All dogs had a diagnosis of appendicular OSA confirmed either by histopathology or cytology with concurrent positive alkaline phosphatase staining.28 In addition to treatment with a standardized palliative protocol (Table 1), the study population of 50 dogs was randomized into 2 groups given either pamidronatea (2.0 mg/kg IV) or placebo (0.9% sodium chloride), in a double-blinded fashion. Pet owners had been informed previously of available treatment options, including surgery with adjuvant systemic chemotherapy, and traditional or investigational palliative therapies, and signed an informed consent form before study entry. Dogs were treated in accordance with the animal care guidelines of the University of Illinois Institutional Animal Care and Use Committee.

Table 1.   Study protocol and population characteristics of 50 dogs with appendicular osteosarcoma treated with standardized palliative therapy with or without pamidronate.
TherapyDay 0Day 7Day 14Day 21Day 28Day 42Day 56Day 63Day 84Q 4 (Weeks)
8 Gy RTH      
Doxorubicin (30 mg/m2)a     
Pamidronate or placebo       
Deracoxib (1–2 mg/kg)bContinuous
CBC     
Chemistry      
Urine NTx     
rBMD     
Gait analysis     
CPIS  
Pamidronate 2.0 mg/kgPlacebo Control
  • a

    Doxorubicin HCl, Ben Venue Laboratories, Bedford, OH.

  • b

    b Deramaxx, Novartis Animal Health, Greensboro, NC.

  • NTx, N-telopeptide; rBMD, relative bone mineral density; CPIS, cumulative pain index score; Dots indicate the actual time for various treatments.

Number of patients26Number of patients24
 Male castrated12 Male castrated14
 Female spayed14 Female spayed10
Mean age (range)8.3 years (4.2–14.0)Mean age (range)8.2 years (5.3–12.3)
Mean weight (range)43.1 kg (25.0–61.2)Mean weight (range)44.3 kg (28.2–76.4)
Breeds Breeds 
 Mix7 Labrador Retriever6
 Rottweiler5 Mix5
 Great Dane3 Rottweiler2
 Labrador Retriever2 Other (1 each)11
 Saint Bernard2  
 Other (1 each)7  
Tumor locations Tumor locations 
 Distal radius13 Distal radius13
 Proximal humerus3 Proximal humerus6
 Proximal femur3 Distal tibia4
 Proximal tibia3 Proximal tibia1
 Distal tibia2  
 Distal femur2  

Inclusion and Exclusion Criteria, Treatment, and Follow-Up

Study candidates were not allowed to have received prior forms of palliative therapy including radiation therapy, cytotoxic chemotherapy, or transdermal analgesics. For inclusion in the study, all dogs had to be randomized with intent-to-treat and receive at least their 1st week of therapy (Day 7, Table 1). Completion of the 84-day study protocol, dictated by 5 consecutive doses of IV doxorubicin, was not necessary to meet inclusion criteria. Doses of pamidronate were diluted to a final volume of 250 mL with 0.9% sodium chloride and administered as a 2-hour constant rate infusion (CRI). The placebo control population of dogs received 0.9% sodium chloride at an equivalent volume and rate of infusion. Dogs determined by their owners' assessment to have achieved some degree of pain alleviation (dichotomous response, “yes” or “no”) received additional treatments according to the study protocol (Table 1). Progression-free interval was calculated as the time from initial therapy (Day 0) until clinical failure, defined as pet owner perceived inadequate pain control or radiographically confirmed pathologic fracture. For dogs experiencing adverse events (AE) unrelated to pain or pathologic fracture that resulted in study discontinuation, progression-free interval was calculated from initial therapy (Day 0) until time of documented AE.

All prospective candidates were required to have baseline clinical staging including CBC, serum biochemistry, urinalysis, 3-view thoracic radiographs, and abdominal ultrasonographic study before entry into the study. If not provided by referring veterinarians, baseline radiographs of the affected limb also were acquired. In addition to routine diagnostic tests, all study entrants also had (1) completion of a subjective pain score (CPIS) by pet owners, (2) quantification of urinary NTx excretion, (3) assessment of rBMD, and (4) assessment of gait with computerized pressure platform gait analysis (Tekscan). Serial evaluations with repeated physical examinations, serum biochemistry, pet owner CPIS acquisition, urine NTx excretion quantification, rBMD calculation, and pressure platform gait analysis were scheduled and performed at the UICCC every 28 days until clinical failure, defined as inadequate pain control or radiographically confirmed pathologic fracture. Repeated imaging studies including limb and thoracic radiographs were performed if deemed clinically indicated by the attending clinician(s). The duration of subjective pain alleviation, reason for study discontinuation, overall survival time, and presumed cause of death were available for all 50 patients as determined from medical records or telephone conversations with referring veterinarians or pet owners.

Subjective Pain Score Evaluation

Pet owners were asked to complete a subjective pain questionnaire comprised of 10 activity and quality-of-life parameters weekly for the 1st 4 weeks, then every 4 weeks thereafter. Each parameter was assigned an integer value ranging from 1 to 4, with ascending integers representing greater degrees of pain or discomfort. Numerical scores for all 10 parameters were summated and expressed as a CPIS, allowing for minimum and maximum total scores of 10 and 40, respectively. Subjective clinical responses to therapy were categorized as improved pain palliation (decreased CPIS) or no improved pain palliation (static or increased CPIS). For the purpose of this study, dogs were considered durable responders if subjective pain alleviation (decreased CPIS) lasted ≥112 days. Dogs that failed to achieve analgesia or experienced pain alleviation <112 days were categorized as nondurable responders.

Urine NTx Excretion

Urine was collected on initial presentation (Day 0) and every subsequent scheduled 28-day recheck visit after the 1st administration of pamidronate. Urine samples (voided or cystocentesis) were immediately centrifuged at 4 °C, 400 × g for 10 minutes, and the supernatant was collected and stored at −20 °C in 2-mL polypropylene cryovials until analysis. Urine NTx concentrations were measured with a commercial ELISAb test kit, previously validated for use in dogs,29,30 and expressed as normalized nanomolar (nM) bone collagen equivalents (BCE) per millimolar (mM) urinary concentration of creatinine.

rBMD of Primary Tumor

At initial presentation and on every subsequent 28-day visit, DEXAc scans were performed to measure rBMD of the primary tumor and of an equivalent anatomical area of the normal contralateral limb. All dogs were sedated and positioned in sternal or lateral recumbency for forelimb or hindlimb lesions, respectively. Sandbags, foam wedges, and tape were used as positioning devices to ensure reproducible and consistent anatomic positioning. Three representative regions of tumor and normal limb were acquired and designated (T1, T2, and T3) and (N1, N2, and N3), respectively. rBMD of the primary tumor was calculated by the following formula:

image

Computerized Pressure Platform Gait Analysis (Tekscan)

At initial presentation and on every subsequent 28-day visit, peak vertical force (PVF) of the diseased leg was measured using a computerized pressure platformd (Tekscan), a 0.4 m × 2 m runway containing 9,152 pressure-sensing cells. Data for all OSA-bearing dogs were collected at a velocity and acceleration representative of a brisk walking gait, with analysis of data points falling within a velocity and acceleration rate of 1.2–1.5 m/s and ± 0.5 m/s2, respectively. Data were collected from 3 separate walking trials per visit, and the PVF of the diseased leg was measured and averaged for all 3 walking trials.

Statistical Analysis

To compare the clinical analgesic activity of a standardized palliative protocol in conjunction with pamidronate or placebo in dogs with appendicular OSA, progression-free intervals (duration of pet owner perceived pain alleviation) was calculated using the Kaplan-Meier method with log rank analysis. Differences in baseline subjective (CPIS) and objective (urine NTx, rBMD, and Tekscan) endpoints between dogs receiving pamidronate or placebo were compared with the Wilcoxon rank-sum test. In dogs receiving pamidronate or placebo, comparison of serial changes in subjective (CPIS) and objective endpoints (urine NTx, rBMD, and Tekscan) were analyzed with repeated measures analysis of variance and post hoc comparisons made with a Tukey-Kramer multiple comparisons test. Finally, in dogs categorized as durable responders (pain alleviation ≥112 days), differences in the percent relative decreases of CPIS and urine NTx, and percent relative increases in rBMD and Tekscan associated with pamidronate or placebo as a function of time (28, 56, 84, and 112 days) were compared with a Wilcoxon rank-sum test. Statistical analyses were performed by a commercial computer software.e Significance was defined as P < .05.

Results

Pamidronate or Placebo Treatments and Associated Outcomes in Dogs with Appendicular OSA

For dogs receiving adjuvant pamidronate (2.0 mg/kg), median progression-free interval for pet owner perceived pain alleviation was 76 days (Fig 1). Forty-six percent (12/26) of dogs completed the 84-day study protocol, and 42% (11/26) of dogs responded subjectively (sustained decrease in CPIS) to adjuvant pamidronate and standardized palliative therapy, achieving durable pain alleviation lasting ≥112 days (median, 168 days; range, 112–279). The cause of death or euthanasia in these 11 durable responder dogs included pathologic fracture (n = 4), progressive pain (n = 3), distant metastases (n = 2), and acute death (n = 2). For the remaining 14 dogs either failing to complete the study protocol or achieve durable analgesia, reasons for study discontinuation included pathologic fracture (n = 6), progressive pain (n = 4), acute increases in serum alanine amino-transaminase activity (n = 2), acute neurologic dysfunction (n = 1), and acute death (n = 1).

Figure 1.

 Kaplan-Meier analysis for the duration of subjective pain alleviation. The number of days in which pet owners perceived their dogs treated with standardized palliative therapy with (solid line) and without (broken line) adjuvant pamidronate to be adequately controlled for pain.

For control dogs receiving saline placebo, median progression-free interval for pet owner perceived pain alleviation was 75 days (Fig 1). Forty-six percent (11/24) of dogs completed the 84-day study protocol, and 38% (9/24) of dogs responded subjectively (sustained decrease in CPIS) to saline placebo and standardized palliative therapy, achieving durable pain alleviation lasting ≥112 days (median, 196 days: range, 112–1,005). The cause of death or euthanasia in these 9 durable responder dogs included pathologic fracture (n = 3), progressive pain (n = 4), distant metastases (n = 1), and acute neurologic dysfunction (n = 1). For the remaining 15 dogs either failing to complete study protocol or achieve durable analgesia, reasons for study discontinuation included pathologic fracture (n = 7), progressive pain (n = 6), distant metastases (n = 1), and acute neurologic dysfunction (n = 1).

Serial Evaluation of Subjective and Objective Markers in Dogs Achieving Durable Pain Analgesia

Based on the subjective characterization of patients being more or less painful as reported by pet owners, no difference in median analgesic duration was demonstrated in dogs receiving standardized palliative therapy with or without adjuvant pamidronate, P= .39 (Fig 1). Although there was no difference in the duration of subjective pain alleviation in the 2 populations of dogs, it remained possible that the degree of analgesia achieved by durable responder dogs could be different between patients treated with or without adjuvant pamidronate. To identify potential clinical or biological differences in the degree of analgesia achieved between durable responders receiving pamidronate (n = 11) or placebo (n = 9), serial changes in subjective and objective surrogates of pain alleviation were evaluated including CPIS, urine NTx, rBMD, and Tekscan. Before comparing serial changes in surrogate parameters, baseline CPIS, urine NTx, rBMD, and Tekscan of both groups (pamidronate or placebo) were compared for similarity (Table 2), and 3 of the 4 surrogate parameters (CPIS, rBMD, and Tekscan) were similar between groups. Basal concentrations of urine NTx, however, were significantly higher in dogs designated to receive pamidronate, suggesting that this group of dogs had greater osteoclastic activity than placebo-treated dogs before study initiation.

Table 2.   Comparison of baseline parameters of 20 durable responder dogs with appendicular osteosarcoma treated with standardized palliative therapy with or without pamidronate.
Baseline ParameterPamidronate
(n = 11)
(Mean ± SEM)
Placebo
(n = 9)
(Mean ± SEM)
P-Value
Cumulative pain score index16.7 ± 0.714.7 ± 0.5.08
Urine N-telopeptide excretion305.7 ± 40.3168.5 ± 33.2.03
Relative tumor bone mineral density1.27 ± 0.151.34 ± 0.16.76
Peak vertical force of diseased limb2.13 ± 0.282.07 ± 0.18.89

Based on completion of a 10-parameter behavioral survey by pet owners, durable responders demonstrated significant decreases in their CPIS in comparison with baseline (Fig 2A). Decreases in CPIS reflect improved behavior and activity levels reported by dog owners after standardized palliative therapy with or without adjuvant pamidronate. For pamidronate-treated dogs, significant CPIS decreases in comparison with baseline were identified at Days 28, 56, and 84 (P < .01 for each respective time point). A similar but shorter lasting pattern was demonstrated in the placebo group, with significant decreases in CPIS from baseline identified at Days 28 and 56 (P < .05 for each respective time point). When comparing changes in CPIS between the 2 treatment groups, the relative percent decrease of CPIS was significantly greater at Day 84 in dogs receiving adjuvant pamidronate, P= .04 (Fig 2B).

Figure 2.

 Serial changes in cumulative pain index score (CPIS) in osteosarcoma-bearing dogs achieving durable pain alleviation. Comparison of CPIS changes relative to baseline (A) every 28 days in durable responder dogs receiving pamidronate (cross-hatch bar) or placebo (open bar). Significant reductions in CPIS from baseline denoted with ♦ (pamidronate) or * (placebo). Differences in the relative percent CPIS reduction (B) reported in dogs treated with pamidronate (cross-hatch) or placebo (open bar) at designated time points. Statistical differenced denoted by *, with significance defined as P < .05.

Significant decreases from baseline urine NTx concentrations were observed at all time points for both pamidronate (P < .001 for all time points) and placebo (P < .01 for Days 28, 56, and 84; P < .05 for Day 112) dogs achieving durable pain alleviation (Fig 3A). Despite baseline urine NTx concentrations being significantly higher in dogs treated with pamidronate (305.7 ± 40.3 nM BCE/mM creatinine) in comparison with placebo (168.5 ± 33.2 nM BCE/mM creatinine), the relative percent decrease in urine NTx was greater at all time points in dogs receiving adjuvant pamidronate, P < .001 for all time points (Fig 3B).

Figure 3.

 Serial changes in urine N-telopeptide (NTx) excretion in osteosarcoma-bearing dogs achieving durable pain alleviation. Comparison of urine NTx changes relative to baseline (A) every 28 days in durable responder dogs receiving pamidronate (cross-hatch bar) or placebo (open bar). Significant reductions in urine NTx from baseline denoted with ♦ (pamidronate) or * (placebo). Differences in the relative percent urine NTx reduction (B) documented for dogs treated with pamidronate (cross-hatch) or placebo (open bar) at designated time points. Statistical differenced denoted by *, with significance defined as P < .05.

For durable responders, increases in primary tumor rBMD in comparison with baseline were observed on Days 56, 84, and 112 for both pamidronate (P < .05 Day 56; P < .01 Day 84; and P < .001 Day 112) and placebo-treated (P < .01 Day 56; P < .05 Day 84; and P < .001 Day 112) dogs (Fig 4A). When comparing changes in primary tumor rBMD between the 2 treatment groups, the relative percent increase in rBMD was significantly greater at Days 84 and 112 in dogs receiving adjuvant pamidronate, P= .04 and .03, respectively (Fig 4B).

Figure 4.

 Serial changes in primary tumor relative bone mineral density in osteosarcoma-bearing dogs achieving durable pain alleviation. Comparison of relative bone mineral density (rBMD) changes relative to baseline (A) every 28 days in durable responder dogs receiving pamidronate (cross-hatch bar) or placebo (open bar). Significant increases in rBMD from baseline denoted with ♦ (pamidronate) or * (placebo). Differences in the relative percent rBMD increases (B) documented for dogs treated with pamidronate (cross-hatch) or placebo (open bar) at designated time points. Statistical difference denoted by *, with significance defined as P < .05.

PVF on the affected limb of OSA-bearing dogs was maintained in patients achieving durable analgesia (Fig 5A), and the administration of pamidronate or placebo did not change PVF as a function of time in durable responding dogs. In dogs prematurely failing palliative therapy (ie, nondurable responders), decreases in PVF in comparison with baseline were readily detectable as early as Day 28, regardless of the treatment group (data not shown). No significant difference in relative percent PVF increase was identified between pamidronate- or placebo-treated dogs achieving durable analgesia (Fig 5B).

Figure 5.

 Serial changes in peak vertical force in osteosarcoma-bearing dogs achieving durable pain alleviation. Comparison of peak vertical force (PVF) changes relative to baseline (A) every 28 days in durable responder dogs receiving pamidronate (cross-hatch bar) or placebo (open bar). Significant increases in PVF from baseline denoted with ♦ (pamidronate) or * (placebo). Differences in the relative percent PVF increase (B) documented for dogs treated with pamidronate (cross-hatch) or placebo (open bar) at designated time points. Statistical differenced denoted by *, with significance defined as P < .05.

Discussion

In this study, the use of adjuvant IV pamidronate did not clinically improve pain alleviation obtained with a standardized palliative protocol combining ionizing radiation, IV doxorubicin, and oral deracoxib. The median progression-free interval for pain alleviation was similar in dogs treated with adjuvant pamidronate or placebo, at 76 or 75 days, respectively. Although pet owners did not directly report enhancement in clinical analgesia provided by adjuvant pamidronate in this study, the greater decreases in urine NTx with concurrent increases in primary tumor rBMD in dogs treated with adjuvant pamidronate support decreased pathologic resorption in the primary tumor bone microenvironment.

Although the duration of pain alleviation was similar in this study for dogs receiving pamidronate or placebo, there is some evidence that the degree of pain alleviation may have been subjectively and objectively different between groups of dogs categorized as durable responders. Dogs receiving adjuvant pamidronate maintained decreases in their CPIS for up to 84 days, whereas placebo-treated dogs only sustained CPIS decreases for 56 days (Fig 2A). Additionally, the relative percent decrease of CPIS from baseline at Day 84 was greater in dogs receiving pamidronate than in placebo-treated patients (Fig 2B). Because CPIS is a composite of 10 questions pertaining to activity and behavior, it may provide more complete information regarding quality-of-life scores in dogs with overtly painful conditions, such as appendicular OSA. Given that pamidronate-treated dogs demonstrated more significant relative percent decreases in CPIS at 84 days in comparison with placebo-treated dogs, pamidronate may provide some added quality-of-life benefits in this subset of patients.

In addition to CPIS, dogs achieving durable analgesia demonstrated differences in urine NTx excretion and primary tumor rBMD. Through the use of inducible murine tumor models, it has been demonstrated that the degree of malignant bone pain is highly correlated with severity of skeletal osteolysis.2,5 As such, quantifying differences in global (urine NTx) and focal (rBMD) bone turnover in OSA-bearing dogs may aid in objectively assessing analgesic responses to palliative therapies.18–20 For urine NTx, both groups of dogs demonstrated significant reductions in NTx concentrations relative to baseline values after institution of standardized palliative therapy (Fig 3A). The observed decrease in urine NTx in dogs receiving placebo strongly suggests that palliative therapy, in exclusion of antiresorptive agents, can decrease osteoclastic activity presumptively within the localized bone tumor microenvironment. Importantly, in dogs receiving adjuvant pamidronate, the relative decrease in urine NTx was greater than that observed in placebo-treated patients at all time points evaluated (Fig 3B). The augmented decrease in urine NTx for pamidronate-treated dogs could result from diminished global homeostatic bone resorption, enhanced decrease in focal osteolysis in the tumor microenvironment beyond that achieved with standard palliative therapy, or a combination of these factors.

Because urine NTx excretion reflects global skeletal turnover, its utility as an independent surrogate endpoint is limited for assessing resorptive changes in focal osteolytic lesions. However, quantifying urine NTx in conjunction with regional imaging studies can provide valuable information pertaining to the local bone tumor microenvironment. Both pamidronate- and placebo-treated dogs experienced increases in rBMD at Days 56, 84, and 112 (Fig 4A). Similar to the pattern observed for urine NTx, the observed increase in rBMD in dogs receiving placebo indicates that the collective effects of ionizing radiation, systemic chemotherapy, and PO administered deracoxib can decrease osteoclastic activity in the regional bone tumor microenvironment. In dogs receiving adjuvant pamidronate, the relative percent increase in primary tumor rBMD was greater than that observed in placebo-treated patients at Days 84 and 112 (Fig 4B). This finding supports the hypothesis that adjuvant pamidronate may provide added beneficial bone biologic effects in the local bone tumor microenvironment.24–26

In this study, a novel objective measure for assessing pain alleviation included the serial evaluation of gait with a computerized pressure platform. Although assessment of gait as a correlate of pain alleviation in OSA-bearing dogs would appear intuitive, this study represents the 1st attempt to utilize and evaluate this objective endpoint in a canine OSA population. In this study, dogs achieving durable analgesia were able to maintain or modestly increase (∼15%) PVF on their diseased leg (Fig 5A), and treatment with pamidronate did not alter relative percent PVF increases in durable responder dogs (Fig 5B). Although minimal fluctuation in PVF (∼5–15%) was identified in dogs categorized as durable responders, patients experiencing progressive pain could be readily identified by pressure platform gait analysis (data not shown). Most dogs failing to achieve clinically relevant pain alleviation demonstrated dramatic reductions (35–75%) in PVF exerted by their diseased leg as early as Day 28 (Fig 6). Collectively, these findings suggest that maintenance or slight improvement of diseased limb usage is the most realistic clinical outcome in OSA-bearing dogs treated with aggressive multimodality palliative therapies.

Figure 6.

 Typical gait analysis image (A) during osteosarcoma (OSA) dog ambulation (right rear [RR], right front [RF], left rear [LR], and left front [LF]). Isolated pictorial changes in PVF exerted by right front legs of 2 individual OSA-dogs achieving poor (B) or good (C) analgesia after 28 days of palliative therapy.

Given that multiple reports describe the individual effectiveness of ionizing radiation or IV aminobisphosphonates for managing bone pain associated with appendicular OSA, it is disappointing that this investigation failed to demonstrate a clear improvement in pet owner perceived pain control for OSA-bearing dogs treated with pamidronate in combination with a standardized palliative protocol. Based upon our current understanding of malignant bone pain generation and positive findings derived from rodent models of malignant osteolysis, it was hypothesized that combining a potent antiosteoclastic agent with ionizing radiation would produce additive analgesic and bone strengthening effects.31,32 Although less pronounced than expected, the findings from our study corroborate reports in people treated with combination radiation therapy and IV pamidronate for bone cancer pain management. Several studies in people suffering from painful skeletal metastases have consistently demonstrated that the addition of IV pamidronate does not increase clinical analgesia beyond what ionizing radiation therapy can provide.33–35 Rather, the justification for adjuvant pamidronate usage in people concurrently treated with radiation therapy is to decrease the risk of malignant skeletal events including pathologic fracture and additional bony metastases.

Although this study provides new information regarding the therapeutic management of malignant bone pain, it had several limitations. First, although this investigation was conducted as a prospective, randomized, double-blind, placebo-controlled study, the number of dogs enrolled was relatively small (n = 50). Therefore, any added benefit exerted by pamidronate would have to be quite large in order to be detected. As such, our conclusion that pamidronate does not provide additional pain alleviation in dogs treated with standardized palliative therapy may be incorrect based on the low power of this study (type II error). Second, the number of pamidronate treatments utilized in this study was restricted to a total of 3 doses administered on Days 0, 28, and 56. Given that people suffering from skeletal metastases receive monthly IV aminobisphosphonates continuously for years to control chronic malignant bone pain, it must be considered that our study design for evaluating the analgesic effect of pamidronate in OSA-bearing dogs may be limited. As such, our negative findings may reflect an error in dosing regimen, rather than an absolute lack of clinical benefit exerted by the drug itself. Finally, although this study includes objective bone metabolic parameters for assessing therapeutic response, the serial changes in urine NTx and rBMD most accurately reflect rates of bone resorption in the local tumor microenvironment, and only provide indirect information regarding pain alleviation. However, based upon established murine models of cancer-induced bone pain in which the degree of osteolysis positively correlates with painful behaviors (limb guarding and wincing), it is reasonable to assume that decreases in focal osteolysis in dogs with OSA would likewise be correlated with some degree of pain alleviation.

Despite the limitations of this study, it provides a clearer perspective on whether adjuvant pamidronate is clinically warranted for the management of painful osteolysis in patients already receiving multimodality palliative therapy. Despite its beneficial bone biologic effects, the results of this investigation fail to strongly support the use of adjuvant pamidronate from a purely analgesic perspective. Additionally, this study introduces an intuitive, yet unexplored, concept that utilizes computerized pressure platform gait analysis as an objective methodology for assessing response to palliative therapy in dogs with painful appendicular OSA lesions. Collectively, this investigation provides key information and validates new tools, which may be useful for evaluating new therapies in managing painful canine neoplastic bone conditions.

Footnotes

aAredia, Novartis Pharma, East Hanover, NJ

bOsteomark NTX urine, Ostex International Inc, Seattle, WA

cQDR-4500W, Hologic, Bedford, MA

dTekscan, Boston, MA

eGraphPad InStat, GraphPad Software Inc, San Diego, CA

Acknowledgments

This study was funded by the Morris Animal Foundation, Grant No. D04CA-109. The authors thank Nancy George for her important role in data collection and management, Kim Knap for her time and effort in computerized pressure platform gait analysis, and Drs Hugues Lacoste, David Heller, Lorin Hillman, Pamela Lucas, Virginia Coyle, Jenny Rose, and Rebecca Moss of the Cancer Care Clinic for patient management.

Ancillary