Histomorphometric and microarchitectural analyses using the 2 mm bone marrow trephine in metastatic breast cancer patients-preliminary results.

BACKGROUND: Bone-targeted agents are widely used for the treatment of osteoporosis, the prevention of cancer-therapy induced bone loss, and for reducing the risk of skeletal related events in patients with metastatic disease. Despite widespread use, relatively little is known about the in vivo effect of these agents on bone homeostasis, bone quality, and bone architecture in humans. Traditionally bone quality has been assessed using a transiliac bone biopsy with a 7 mm "Bordier" core needle. We examined the possibility of using a 2 mm "Jamshidi" core needle as a more practical and less invasive method to assess bone turnover and potentially other tumor effects.
METHODS: A pilot study on the feasibility of assessing bone quality and microarchitecture and tumor invasion using a 2 mm bone marrow trephine was conducted. Patients underwent a posterior trans-iliac trephine biopsy and bone marrow aspirate. Samples were analyzed for bone microarchitecture, bone density, and histomorphometry. The study plan was to accrue three patients with advanced breast cancer to assess the feasibility of the study before enrolling more patients.
RESULTS: The procedure was well tolerated. The sample quality was excellent to analyze bone trabecular microarchitecture using both microCT and histomorphometry. Intense osteoclastic activity was observed in a patient with extensive tumor burden in bone despite intravenous bisphosphonate therapy. DISCUSSION: Given the success of this study for assessing bone microarchitecture, bone density, and histomorphometry assessment using a 2 mm needle the study will be expanded beyond these initial three patients for longitudinal assessment of bone-targeted therapy.

Introduction

Bone-targeted agents aimed at reducing osteoclastogenesis including bisphosphonates and <span class="Chemical">denosumab are widely used in cancer patients. Their uses include; the treatment of osteoporosis, prevention of cancer-therapy induced bone loss, reducing the risk of skeletal related events, reducing pain, and improving quality of life in patients with metastatic disease [1]. In cancer patients these agents are administered systemically at high and frequent, often monthly, doses and for extended periods of time relative to their use in the treatment of osteoporosis. Despite this there is a paucity of data about the in vivo effect of these agents on bone homeostasis, bone quality, and microarchitecture in humans. This topic has gained increasing attention recently with the awareness of long term toxicities of bisphosphonates such as osteonecrosis of the jaw (ONJ) and atypical fractures.[2], [3].

Traditionally bone strength has been viewed as an integration of bone quantity and bone quality [4]. Clinically, bone strength and susceptibility to fracture is based on bone mineral density (BMD) as reflected through dual energy X-ray absortiometry (DEXA) scans [5]. This, however, has only modest sensitivity/specificity [5], [6], and only a small proportion of fracture risk reduction is actually explained by bone density increases [7], [8].

Clinical and laboratory evidence suggest mechanical properties, in addition to BMD, play an important role in bone strength [9], [10], [11]. However, assessing bone quality requires invasive testing, traditionally with a 7–8 mm “Bordier” trephine of transiliac bone [12]. The size of the needle poses two problems. First, it is uncomfortable and painful for the patient. Second, it is difficult technically in achieving an adequate sample in cancer patients on bone-targeted agents as anecdotally the procedure is more challenging than in patients with osteoporosis [13]. Previous data from cadavers has demonstrated that the 2 mm trephine provides comparable results to a 7 mm trephine [14]. A review of the literature revealed no studies comparing the two trephine sizes in living cancer patients.

Given the widespread use of outpatient posterior iliac crest bone marrow biopsies in hematology and oncology using a 2 mm Jamshidi needle, we wanted to see if this would allow a more practical, technically easier sampling a<span class="Species">menable to routine screening of bone status in cancer patients.

Methods

Subjects and design

This study was a pilot feasibility study performed in three patients prior to starting a larger prospective, single-arm, non-interventional feasibility study using a 2 mm <span class="Chemical">trephine to assess bone quality in breast cancer patients. Patients were required to have: advanced breast cancer, with or without bone metastases with or without bisphosphonate treatment, and an ECOG ≤2. All patients enrolled underwent a bone mineral density assessment using DEXA. The study was approved by the research ethics committee at the Ottawa Hospital Cancer Centre, and all patients provided written informed consent.

Bone specimen collection

Trans-iliac crest bone biopsy specimens were obtained from the posterior iliac crest using a Jamshidi bone biopsy <span class="Chemical">trephine (diameter 2 mm) (Cardinal Health, Dublin, Ohio, United States of America) (see Appendix 1). The samples underwent histomorphometric analysis, microarchitectural analysis, and routine pathologic assessment.

Histological and histomorphometrical analysis of bone

Hematoxyline and Eosin (HE), <span class="Disease">von Kossa and van Gieson (VKVG), and tartrate-resistant acid phosphatase (TRAP) staining of plastic sections of posterior iliac bone samples from three patients were performed on plastic sections. As described elsewhere, for plastic sectioning, bone samples fixed in 4% PFA/PBS were embedded in methyl methacrylate, and sectioned (7-μm thickness), and von Kossa and van Gieson staining was applied [15]. Unmineralized bone sections were analyzed using Osteomeasure software (Osteometrics, Inc.). Images were taken at room temperature using a light microscope (DM200; Leica) with a 20×(numerical aperture of 0.40) or 40×(numerical aperture of 0.65) objective. All histological images were captured using a camera (DP72; Olympus), acquired with a DP2-BSW software (XV3.0; Olympus), and processed using Photoshop (Adobe).

Three dimensional micro-computed tomography (3D microCT) of bone samples

The core biopsy samples were scanned wet in 70% ethanol by micro-CT at 40X magnification with a SkyScan 1072 (Skyscan, Aartselaar, Belgium) and analyzed with a bone analysis software (ver2.2f, Aartselaar, Belgium). Trabecular bone structure was measured. Parameters were acquired with a rotation of 0.9° between each picture and the x-ray source set at 100 kV and 98 μA. The seg<span class="Species">mentation of the image was made by a global threshold and a voxel size of 21.90×21.90×21.90 μm³; the same threshold setting was used for all the samples. Architectural measurements were made as previously described [16], [17].

Bone mineral density measurements by DEXA

Core biopsies were subjected to bone mineral density (BMD) analysis using a PIXIMUS bone densitometer (PIXIMUS TM, GE medical systems, Schenectady, NY, USA). A quality control phantom was used to calibrate the densitometer prior to each experiment. All patients also underwent DEXA scanning.

Statistical analysis

The primary endpoint of this pilot study was to assess the feasibility of obtaining sufficient tissue using bone marrow aspirates and trephine biopsy for histomorphometry. The secondary endpoint was comparison between current bone mineral density and pathological analysis of bone marrow <span class="Chemical">trephine biopsy looking at histomorphometry, and micro-architecture analysis. Given this was a pilot study simple descriptive statistics were used.

Results

Between January and July 2011 three patients consented and underwent outpatient biopsies. Patient number one and number two received two biopsies from separate areas of the posterior iliac crest. Patient three received a single biopsy. The demographic characteristics of each patient can be found in Table 1. Notably, all patients were post-menopausal, had metastatic disease, and two were on long-term bisphosphonate therapy.

Table 1

Demographic data of study patients.

	Patient 1	Patient 2	Patient 3
Age	64	56	59
Menopause status	Postmenopausal	Postmenopausal	Postmenopausal
BMI (kg/m²)	34.0	19.5	27.9
Smoking history	None	None	None
BMD (t-score)	−1.9 [L1–L2]	−1.3 [L1–L4]	−2.2 [L1–L4]
Breast cancer	Metastatic to soft tissues	Extensive bone metastases	Limited bone metastases
Current therapy	Aromatase inhibitor	Aromatase inhibitor	Aromatase inhibitor
Bisphosphonate use and duration	Nil	Yes (3 years)	Yes (2 years)

BMI=body mass index, BMD= bone mineral density and L=lumbar spine (i.e. L1=first lumbar vertebrae).

3D microCT data

Bone volume (BV/TV) ranged from 10.8 to 13.5% consistent with the published literature [18]. Trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), trabecular number (Tb.N) and bone surface per tissue area (BS/TV) values were also consistent with the range published in the literature in post-<span class="Species">menopausal women [18]. 3D microCT data are located in Table 2. The accompanying figures for each biopsy are located in Fig. 1.

Table 2

Microarchitectural parameters and measurement of cellular activities in the poster iliac crest assessed by histomorphometry and 3D microCT.

Sample ID	Patient1	Patient 2	Patient 3
BV/TV (%)	10.9	13.5	10.8
Tb. Th (mm)	0.14814	0.173	0.132
Tb. Sp (mm)	0.726	0.570	0.575
Tb.N (mm)	0.741	0.781	0.819
Ct. Wi (μm)	N/A	N/A	N/A
BS/TV (1/mm)	2.707	3.049	2.98

BV/TV: bone volume.

Tb.Sp: trabecular separation.

Ct.Wi: cortical width.

N/A: non-applicable.

Tb.Th: trabecular thickness.

Tb.N: trabecular number.

BS/TV: bone surface per tissue area.

Fig. 1

MicroCT images, for patient #1, #2, and #3.

Bone histomorphometry

Hematoxylin–Eosin (HE) stain of the hisotology sections shows the presence of tumor (T) in sample of <span class="Species">patient 1 and 2 (Fig. 2). Samples from patient 1 and patient 2 shows the presence of osteoid (O) by von Kossa and van Gieson (VKVG) stain. Among the microscopic field presented only sample from patient 2 shows the presence of osteoclasts (C) as evident by tartrate-resistant acid phosphatase (TRAP) staining (third panel). Osteoblasts were stained with toludine blue. Osteoblast and osteoclast numbers per tissue area and bone perimeter are displayed in Table 3.

Fig. 2

Histological and histomorphometric analysis of bone. Hematoxyline and Eosin (HE), von Kossa and van Gieson (VKVG), and tartrate-resistant acid phosphatase (TRAP) staining of plastic sections of bone samples from three patients. HE stain shows the presence of tumor (T) in patient #2. Patient #1 and #2 shows the presence of osteoid volume (O) by VKVG stain (middle panel). No osteoclast is seen in the field shown for sample of patient #1. BM: bone marrow and MB: mineralized bone. Among the microscopic field presented only sample from patient #2 show the presence of osteoclasts (C) by TRAP stain (right panel). Osteoblasts were stained with toludine blue.

Table 3

Osteoblast and osteoclast count data.

Patient	N.Ob/T.Ar	N.Ob/B.Pm	N.Oc/T.Ar	N.Oc/B.Pm
1	142.57	14.55	2.22	0.22
2	142.57	16.98	6.68	0.80
3	100.25	10.91	0	0

N.Ob/T.Ar: number of osteoblasts/ tissue area N.Ob/B.PM: number of osteoblasts/bone perimeter Oc/T.Ar: number of osteoclasts/ tissue area and N.Oc/B.PM: number of osteoclasts/bone perimeter in different patient bone sections.

Discussion

Despite the widespread use of bone-targeted agents in breast cancer <span class="Species">patients, there is a relative paucity of data about the in vivo effect of these agents on bone homeostasis, bone quality, and microarchitecture in humans [1]. This is not withstanding that we are using increasingly potent agents at higher doses and increased frequencies, relative to their use in the treatment of osteoporosis. This is important as long-term bisphosphonate use confers prolonged suppression of bone remodeling sites which can have deleterious clinical effects [3], [19], [20], [21], [22].

Newer methods for assessing bone quality are under investigation, such as microindentation technique and high-resolution peripheral quantitative CT [23], [24], [25], but are regarded as exploratory. High resolution-peripheral quantitative CT appears to be better than DEXA at predicting <span class="Disease">fracture risk, but has not yet reached widespread clinical use [24], [25], [26], [27]. Further, the use of biochemical turnover markers [8] are primarily used for monitoring treatment efficacy and compliance, and do not directly assess bone quality. Bone histomorphometry remains the gold standard in evaluating bone status prior to and following therapeutic interventions. However, this invasive procedure is not amenable to routine use. In cancer patients bone marrow biopsy is frequently used for pathological assessment and we explored the possibility to extend its use to bone histomorphometric parameters. We therefore assessed the feasibility of using 2 mm core biopsies to obtain additional information on in vivo bone biology indices.

The current study demonstrates that using a 2 mm Jamshidi is feasible, practical and yields a sample sufficient to determine trabecular microarchitecture and bone turnover indices by both microCT and histomorphometry. The benefit of this technique is that it provides qualitative information on the in vivo effect of bone-targeted agents on bone quality.This technique may also provide a feasible means of assessing bone quality in cancer <span class="Species">patients on therapies known to affect bone quality (e.g., tamoxifen, aromatase inhibitors). Previous studies have used BMD in this setting, but it has its limitations and is not able to assess bone quality [28], [29], [30]. The importance of assessing bone quality and balancing the potential risks against the benefits of bone toxic agents will be of paramount importance if such agents (e.g., exemestane) are used for primary prevention of breast cancer [23], [31].

The findings of this prospective, pilot study are limited by the small sample size, but recruitment is currently under way to expand the study cohort. Another important limitation is that preclinical models have demonstrated that the effect of <span class="Chemical">bisphosphonates on bone turnover is highly site specific [32]. The implications are that data from a single biopsy site (i.e. iliac crest) may not be representative of global bone quality [32], [33]. There were, however, several findings that merit consideration. Patient #1 had metastatic cancer to soft tissues but no bone metastasis and was treated with aromatase inhibitors but not with bisphosphonates. Bone histomorphometry revealed active bone turnover as indicated in Table 3. In contrast patient #3 who had limited bone metastatic disease was treated with both aromatase inhibitors and bisphosphonates and had evidence of bone suppression with no active osteoclast detected. Patient #2 was of particular interest because of extensive bone metastasis by skeletal survey and its confirmation by the presence of metastatic tumor on histomorphometric sections (Fig. 2). Interestingly this patient had very active osteoclastic activitiy as indicated in Table 3 despite treatment with intravenous bisphosphonates for three years suggesting that bisphosphonates were unable to completely suppress turnover when tumor burden is high. Given our sample size, it is premature to generalize our findings but it provides a rationale to extend our study with a larger sample size to assess the reproducibility of our findings and whether or not a 2 mm core will be able to replace the use of 7 mm core biopsies.

Conclusion

The 2 mm Jamshidi is feasible, practical, and yields an adequate sample to assess bone structure and quality using microCT and histomorphometry in breast cancer patients. Additional samples are required to assess reproducibility, effect of bisphosphonates on bone quality and novel information on bone/tumor interaction. Further research is required to determine whether this technique may be to longitudinally assess bone quality in cancer patients on long-term bone-targeted therapies.