Method for predicting bone damage reversal

Abstract

A method for predicting the extent of bone-building drug-or bone-building therapy-induced reversal of bone damage in a subject who has been administered at least one bone-building drug and / or has undergone bone-building therapy, the method comprising the steps of: (a) obtaining a keratinized tissue sample from the subject; (b) obtaining a Raman spectrum of the keratinized tissue sample; (c) analysing the Raman spectrum to identify at least one marker indicative of bone-building drug- or therapy-induced reversal of bone damage; and (d) predicting the extent of bone-building drug- or therapy-induced reversal of bone damage based on the identification of the at least one marker in the Raman spectrum.

Description

[0001] METHODS FOR PREDICTING BONE DAMAGE REVERSAL

[0002] Technical Field of the Invention

[0003] The present invention relates to the use of Raman spectroscopy to predict the extent of bone-building drug- or bone-building therapy-induced reversal of bone damage.

[0004] Background to the Invention

[0005] Osteoporosis is characterised by a drop in bone mineral density (BMD), which can lead to increased susceptibility to fracture. The disease affects both men (approximately one fifth lifetime risk) and women (approximately one third lifetime risk), with incidence typically increasing with age. The highest group at risk from osteoporosis is postmenopausal women, because the decline in oestrogen (E2) levels after menopause triggers severe bone loss.

[0006] Chief amongst therapies for improving the outcomes of patients with osteoporosis are bisphosphonates, a family of compounds based on an analogue of pyrophosphate, where the central oxygen has been replaced by a carbon that can be functionalised with two side groups. The pharmacological action of all bisphosphonates is considered to be similar; they adsorb strongly to the hydroxyapatite component of bone, binding to it and preventing further degradation. The main consequences of the reduction in bone turnover are increased mineralisation, accumulation of microdamage, and a change in collagen cross-linking.

[0007] BMD measured by dual energy X-ray absorptiometry is the gold standard for clinical diagnosis of osteoporosis. However, it is widely understood that the ability of bone to withstand physical stress and resist fracture depends not only on BMD but also other factors including degree of mineralisation, micro-damage, and the organisation of collagen fibres mediated by the type and extent of collagen cross-links. Indeed, BMD is reported as being a poor predictor of fracture and very limited in its ability to monitor response to treatment.

[0008] There is a need for improved and simpler methods for monitoring response to bonebuilding drug / therapy-based treatment, and in particular, for predicting the extent of bone-building drug- or bone-building therapy-induced reversal of bone damage.

[0009] It is an aim of embodiments of the present invention to overcome or mitigate at least one problem of the prior art, whether expressly described herein or not.

[0010] Summary of the Invention

[0011] According to a first aspect of the invention, there is provided a method for predicting the extent of bone-building drug- or bone-building therapy-induced reversal of bone damage in a subject who has been administered at least one bone-building drug and / or has undergone bone-building therapy, the method comprising the steps of:

[0012] (a) Obtaining a keratinized tissue sample from the subject;

[0013] (b) Obtaining a Raman spectrum of the keratinized tissue sample;

[0014] (c) Analysing the Raman spectrum to identify at least one marker indicative of bonebuilding drug- or therapy -induced reversal of bone damage; and

[0015] (d) Predicting the extent of bone-building drug- or therapy-induced reversal of bone damage based on the identification of the at least one marker in the Raman spectrum.

[0016] Surprisingly, the present inventors have found that Raman spectroscopy performed on a keratinized tissue sample obtained from a subject can be used to effectively predict the extent of bone-building drug- or bone-building therapy-induced reversal of bone damage in the subject. Without wishing to be bound by theory, Raman spectroscopy performed on the keratinized tissue sample has been found to be capable of successfully detecting reorganisation of the protein structure of the keratinized tissue (and even highly subtle changes in reorganisation of the protein structure), which has been found to be an accurate indicator of the reversal of bone damage in the subject as a result of bonebuilding drug / therapy-based treatment.

[0017] At least one bone-building drug may be independently selected from the group consisting of: a bisphosphonate, a selective estrogen receptor modulator (SERM), calcitonin, a monoclonal antibody, a parathyroid hormone analog, a selective estrogen receptor modulator (SERM) with estrogen agonist / antagonist properties, calcium and / or vitamin D supplements, and combinations thereof.

[0018] In preferred embodiments, at least one bone-building drug comprises at least one bisphosphonate.

[0019] At least one bisphosphonate may be independently selected from the group consisting of: alendronate, risedronate, ibandronate, zoledronic acid, etidronate, clodronate, pamidronate, tiludronate, and combinations thereof. In some preferred embodiments, at least one bisphosphonate is independently selected from the group consisting of: alendronate, risedronate, ibandronate, and combinations thereof.

[0020] In some embodiments, bone-building therapy may be independently selected from the group consisting of: hormone-replacement therapy (HRT), physical therapy, bone- targeted therapy, and combinations thereof. Bone damage may be independently selected from the group consisting of: bone bruises, bone spurs, bone cysts, bone dislocations, bone degeneration, bone deformities, bone inflammation, bone necrosis, and combinations thereof.

[0021] Bone damage may be due to a bone-related disease or condition that is independently selected from the group consisting of: a fracture, a stress fracture, osteoporosis, osteoarthritis, rheumatoid arthritis, Paget’s disease, osteomyelitis, avascular necrosis, bone cancer, and combinations thereof. In some preferred embodiments, the bone damage may be due to osteoporosis.

[0022] The subject may preferably be a mammal. In particularly preferred embodiments, the subject is a human.

[0023] The subject may be a male or a female, preferably a human male or human female.

[0024] In some embodiments, the subject may be a post-menopausal human female.

[0025] The subject may have a bone-related disease or condition. The subject may have a bone- related disease or condition that is independently selected from the group consisting of: a bone fracture, osteoporosis, osteoarthritis, rheumatoid arthritis, Paget’s disease, osteogenesis imperfecta, osteomalacia, scoliosis, kyphosis, lordosis, bone cancer, fibrous dysplasia, osteochondritis dissecans, avascular necrosis, osteomyelitis, and combinations thereof. In some preferred embodiments, the subject may have osteoporosis.

[0026] In some embodiments, the subject may be a human with an age of at least 10 years, or at least 20, 30, 40, or at least 45 years, or at least 50, 55, or at least 60 years. In some embodiments, the subject may be a human with an age of no greater than 100 years, or no greater than 95, 90, or no greater than 85 years. The keratinized tissue sample may comprise any biological sample that comprises the protein keratin. The keratin may be hard keratin.

[0027] The keratinized tissue sample may be independently selected from the group consisting of: a nail sample, a hair sample, a skin sample, and combinations thereof.

[0028] In preferred embodiments, the keratinized tissue sample is a nail sample. The nail sample may be a fingernail sample and / or a toenail sample.

[0029] The nail sample may be a clipped nail sample. Alternatively, the nail sample may still be attached to the body of the subject and not removed prior to obtaining a Raman spectrum of the sample in step (c).

[0030] The keratinized tissue sample may be obtained by any clinically acceptable methods known in the art. In some embodiments, obtaining the keratinized tissue sample from the subject may comprise clipping a fingernail or toenail from the subject. The fingernail or toenail may be clipped from a free edge of a nail plate of the subject.

[0031] In some embodiments, step (b) comprises obtaining the Raman spectrum of the keratinized tissue sample in situ (i.e. the keratinized tissue sample may be still attached to the body of the subject). In other embodiments, step (b) comprises obtaining the Raman spectrum of a collected keratinized tissue sample that is not attached to the body of the subject.

[0032] The Raman spectrum may be obtained using any Raman spectroscopy technique or any technique that produces Raman spectra. The Raman spectroscopy technique may be independently selected from the group consisting of: back- scattering Raman spectroscopy, offset Raman spectroscopy (SORS), Transmission Raman spectroscopy, Coherent Anti-Stokes Raman spectroscopy (CARS), Surface Enhanced Raman spectroscopy (SERS), and Resonance Raman spectroscopy (UV or visible).

[0033] The Raman spectrum may be obtained using a Raman spectroscopy system comprising a laser light source, a spectrometer, and a detector configured to analyse scattered light from the keratinized tissue sample. The keratinized tissue sample may be irradiated by a light source, such as a laser, and the wave number and intensity of inelastically scattered light may be measured.

[0034] The Raman spectrum of the keratinized tissue sample may be obtained at a depth of at least 20 microns into the keratinized tissue sample, or at least 25, 30, 35, 40, or at least 45 microns into the sample. The Raman spectrum of the keratinized tissue sample may be obtained at a depth of no greater than 90 microns into the sample, or no greater than 85, 80, 75, 70, 65, 60, or no greater than 55 microns into the sample. The Raman spectrum of the keratinized tissue sample may be obtained at a depth of between 30-70 microns into the sample, or between 40-60 microns into the sample.

[0035] The obtained Raman spectrum may cover a spectral region between 300 cm1and 1800 cm1, or between 400-1800 cm1.

[0036] In some embodiments, the keratinized tissue sample is obtained from the subject and / or the Raman spectrum of the keratinized tissue sample is obtained at least 1 week after the subject was first administered bone -building drugs and / or first underwent bone-building therapy, or at least 2 weeks, 3 weeks, or at least 1 month, 2 months, 3 months, 4 months, 5 months, or at least 6 months after the subject was first administered bone-building drugs and / or first underwent bone -building therapy. In some embodiments, the keratinized tissue sample is obtained from the subject and / or the Raman spectrum of the keratinized tissue sample is obtained between 1-24 months after the subject was first administered bone-building drugs and / or first underwent bone-building therapy, or between 3-12 months, or between 6-12 months after the subject was first administered bone-building drugs and / or first underwent bone-building therapy.

[0037] The Raman spectrum of the keratinized tissue sample may be obtained immediately after obtaining the tissue sample from the subject. In some embodiments, the keratinized tissue sample may be obtained from the subject and the Raman spectrum of the tissue sample may be taken a period of time after obtaining the sample. The period of time may be at least 20 minutes after obtaining the sample or at least 40 minutes after obtaining the sample, or at least 1 hour, 2, 3, 4, 5, 10, 15, or at least 20 hours after obtaining the sample, or at least 1 day after obtaining the sample. The period of time may be no greater than 1 week after obtaining the sample, or no greater than 6 days, 5, 4, 3, 2, or no greater than 1 day after obtaining the sample, or no greater than 20 hours, 15, 10, or no greater than 5 hours after obtaining the sample.

[0038] In some embodiments, step (c) comprises subjecting the Raman spectrum to a multivariate analysis technique to identify at least one marker indicative of bone-building drug- or therapy -induced reversal of bone damage. The multivariate analysis technique may comprise a chemometric technique. The multivariate analysis technique may be independently selected from the group consisting of: principal component analysis (PCA), partial least squares (PLS) regression, and a combination thereof. The multivariate analysis technique may preferably comprise partial least squares (PLS) regression. The multivariate analysis technique may be used to remove background spectral features from the obtained spectrum. In some embodiments, step (c) comprises the step of comparing the obtained Raman spectrum with at least one reference spectrum or reference spectral data to identify at least one marker indicative of bone-building drug- or therapy -induced reversal of bone damage.

[0039] At least one reference spectrum or reference spectral data may be obtained from a keratinized tissue sample of a subject that does not have damaged bones and / or from a subject who has damaged bones. The keratinized tissue sample of the subject who has damaged bones may be obtained from the same subject on which the method of the invention is performed, preferably prior to the administration of bone-building drugs and / or therapy. The keratinized tissue sample of the subject who does not have damaged bones may be obtained from a different subject, preferably a healthy subject. The obtained Raman spectrum may be compared with both a reference spectrum or reference spectral data obtained from a keratinized tissue sample of a subject that does not have damaged bones and from a subject who has damaged bones.

[0040] At least one marker may show a physical and / or chemical change in the keratinized tissue sample compared to a sample from a subject who has damaged bones (such as a sample obtained from the same subject before onset of drug / therapy-based treatment) that is indicative of bone-building drug- or therapy-induced reversal of bone damage. At least one marker may show a physical and / or chemical change in the keratin present in the keratinized tissue sample.

[0041] At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the obtained Raman spectrum of the keratinized tissue. Said change in a peak may be independently selected from the group consisting of: a newfound presence of the peak in the obtained spectrum compared to a reference spectrum or reference spectral data, disappearance of the peak in the obtained spectrum compared to a reference spectrum or reference spectral data, increase in intensity of the peak in the obtained spectrum compared to a reference spectrum or reference spectral data, decrease in intensity of the peak in the obtained spectrum compared to a reference spectrum or reference spectral data, change in peak sharpness in the obtained spectrum compared to a reference spectrum or reference spectral data, shift in the wavenumber of the peak in the obtained spectrum compared to a reference spectrum or reference spectral data, and combinations thereof.

[0042] Changes in peak intensity may be measured using a technique that is independently selected from the group consisting of: comparison of a width at half maximum of a peak, comparison of a relative peak height, and comparison of an area under a peak (integration), and combinations thereof.

[0043] At least one marker comprising a change relative to a reference spectrum or reference spectral data in one or more peaks may denote a change in the chemical structure of at least one protein in the keratinized tissue sample. Said protein preferably comprises keratin. The change in the chemical structure of at least one protein, preferably keratin, in the keratinized tissue sample may be indicative of bone-building drug- or therapy- induced reversal of bone damage.

[0044] At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the obtained Raman spectrum of the keratinized tissue sample that denote the level of sulfur bonding in the keratinized tissue sample. At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks denoting the extent of cysteine oxidation in the keratinized tissue sample. At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks denoting the extent of disulfide bond and / or bridge formation between cysteine molecules in the keratinized tissue sample.

[0045] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks denoting cysteinecysteine S-S bond stretching. At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks denoting cysteinecysteine S-S bond stretching that are independently selected from the group consisting of: the gauche-gauche-gauche (GGG) conformation of cysteine-cysteine S-S bond stretching, the gauche-gauche-trans (GGT) conformation of cysteine-cysteine S-S bond stretching, and combinations thereof.

[0046] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks denoting cysteine- cysteine S-S bond stretching in the GGG conformation. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 500-510 cm1, or between 500-505 cm1, or preferably between 500-503, or between 501-503 cm1.

[0047] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks denoting cysteine- cysteine S-S bond stretching in the GGT conformation. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 510-520 cm1, or between 515-520 cm1, or preferably between 515-519, or between 518-519 cm1. At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks denoting cysteine CH-CO2 stretching, which may be orthorhombic stretching. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 630-640 cm1, or between 635-640, or between 638-639 cm1.

[0048] At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks denoting cysteine C-S stretching, which may be orthorhombic stretching. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 690-670 cm1, or between 690-696, or between 693-696 cm1.

[0049] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the obtained Raman spectrum of the keratinized tissue sample that denote an a-helix secondary structure in the keratinized tissue sample.

[0050] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the obtained Raman spectrum of the keratinized tissue sample that denote a P-sheet secondary structure in the keratinized tissue sample.

[0051] At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks denoting C-C stretching in an a-helix secondary structure. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 920-950 cm1.

[0052] At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks denoting C-C stretching in a random protein secondary structure. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 960-970 cm1.

[0053] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks denoting C-C stretching in a P-sheet secondary structure. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 970-1010 cm1, or between 990-1005 cm1.

[0054] At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks denoting phenylalanine ring breathing. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 1001-1005 cm1, or between 1002-1003 cm1.

[0055] In some embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks within the amide III range of the Raman spectrum. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 1200-1350 cm1, or between 1220-1320 cm1.

[0056] At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks within the amide III spectral region and denoting a P- sheet secondary structure. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 1230-1242 cm-1.

[0057] At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks within the amide III spectral region and denoting a random secondary structure. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 1245-1255 cm-1. In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks within the amide III spectral region and denoting an a-helix secondary structure. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 1265-1320 cm1.

[0058] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks within the amide III spectral region and denoting a globular a-helix secondary structure. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 1265-1285 cm1.

[0059] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks within the amide III spectral region and denoting a fibrous a-helix secondary structure. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 1285-1320 cm1.

[0060] In some embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks within the amide I range of the Raman spectrum. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 1645-1680 cm1.

[0061] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks within the amide I spectral region and denoting an a-helix secondary structure. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 1645-1660 cm1.

[0062] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks within the amide I spectral region and denoting a P-sheet secondary structure. Said peak or peaks may be located in the Raman spectrum at a wavenumber of between 1670-1680 cm1.

[0063] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 400 and 700 cm1, or between 500-600 cm1. Peaks in such a range may relate to cysteine amino acids in the keratinized tissue sample.

[0064] At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 920 and 950 cm1. Peaks in such a range may relate to an a-helix secondary structure in the keratinized tissue sample.

[0065] At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 960 and 970 cm1. Peaks in such a range may relate to a C-C stretch in a random secondary structure in the keratinized tissue sample.

[0066] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 970-1010 cm1, or between 990-1005 cm1. Peaks in such a range may relate to a P-sheet secondary structure in the keratinized tissue sample.

[0067] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 1200-1350 cm1, or between 1220-1320 cm1. Peaks in such a range may be within the amide III range of the Raman spectrum. At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 1230-1242 cm1. Peaks in such a range may be amide III peaks for a P-sheet secondary structure in the keratinized tissue sample.

[0068] At least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 1245-1255 cm1. Peaks in such a range may be amide III peaks for a random secondary structure in the keratinized tissue sample.

[0069] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 1265-1320 cm1. Peaks in such arange may be amide III peaks for an a- helix secondary structure in the keratinized tissue sample.

[0070] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 1265-1285 cm1. Peaks in such a range may be amide III peaks for a globular a-helix secondary structure in the keratinized tissue sample.

[0071] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 1285-1320 cm1. Peaks in such a range may be amide III peaks for a fibrous a-helix secondary structure in the keratinized tissue sample.

[0072] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 1645-1680 cm1. Peaks in such a range may be within the amide I range of the Raman spectrum.

[0073] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 1645-1660 cm1. Peaks in such a range may be amide I peaks for an a- helix secondary structure in the keratinized tissue sample.

[0074] In some preferred embodiments, at least one marker may comprise a change relative to a reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 1670-1680 cm1. Peaks in such a range may be amide I peaks for a P- sheet secondary structure in the keratinized tissue sample.

[0075] Analysis of the identified markers indicative of bone-building drug- or therapy-induced reversal of bone damage may be used to predict whether reversal of bone damage has occurred. In some embodiments, analysis of the identified markers may be used to quantitatively predict the extent of bone damage reversal.

[0076] In some embodiments, bone damage reversal may comprise a change in the protein structure of bone, which may comprise a change in the collagen structure of bone. The change in the protein structure of bone may comprise a reversion of protein structure disruption. The change in the protein structure of bone may comprise reorganisation of the protein structure, which may comprise a reorganisation from a sheet-based protein structure to a more fibrous structure. In some embodiments, bone damage reversal may comprise a reversal of osteoporosis or symptoms thereof.

[0077] According to a second aspect of the invention, there is provided a method for predicting the extent of bone-building drug- or bone-building therapy-induced reversal of bone damage in a subject who has been administered at least one bone-building drug and / or has undergone bone-building therapy, the method comprising the steps of:

[0078] (a) Obtaining a keratinized tissue sample from the subject;

[0079] (b) Obtaining a Raman spectrum of the keratinized tissue sample;

[0080] (c) Comparing the obtained Raman spectrum of the keratinized tissue sample with at least one reference spectrum or with reference spectral data; and

[0081] (d) Predicting the extent of bone-building drug- or therapy-induced reversal of bone damage based on the results of the comparison of the obtained Raman spectrum with the at least one reference spectrum or with the reference spectral data.

[0082] The method of the second aspect of the invention may be the method of the first aspect of the invention. Statements of invention for the first aspect of the invention above may also be applied mutatis mutandis to the second aspect of the invention.

[0083] According to a third aspect of the invention there is provided a method of monitoring bone-building drug- or bone-building therapy-induced reversal of bone damage over time in a subject who has been administered at least one bone-building drug and / or has undergone bone-building therapy, the method comprising the steps of:

[0084] (a) Obtaining multiple keratinized tissue samples from the subject over a period of time;

[0085] (b) Obtaining a Raman spectrum of each obtained keratinized tissue sample;

[0086] (c) Analysing each Raman spectrum to identify at least one marker indicative of bone-building drug- or therapy-induced reversal of bone damage; and (d) Monitoring changes in the at least one identified marker over time to assess the progression of bone -building drug- or therapy-induced reversal of bone damage.

[0087] Statements of invention for the first aspect of the invention above may also be applied mutatis mutandis to the third aspect of the invention. Statements of invention below for the third aspect of the invention may also be applied mutatis mutandis to the other aspects of the invention.

[0088] The method may monitor bone-building drug- or bone-building therapy-induced reversal of bone damage over a total time period of at least 3 months, or at least 4, 5, 6, 7, 8, 9, 10, 11, or at least 12 months, or at least 15, 18, 21, or at least 24 months.

[0089] Keratinized tissue samples may be obtained from the subject over a total time period of at least 3 months, or at least 4, 5, 6, 7, 8, 9, 10, 11, or at least 12 months, or at least 15, 18, 21, or at least 24 months.

[0090] In some embodiments, a keratinized tissue sample may be obtained from the subject at least once every 1 year, or at least once every 9 months, or at least once every 6 months, or at least once every 3 months, or at least once every 1 month.

[0091] In some embodiments, the step of monitoring changes in at least one identified marker over time comprises monitoring changes in one or more peaks in the obtained Raman spectra of the keratinized tissue samples over time. Said changes in a peak may be independently selected from the group consisting of: a new peak forming in the Raman spectra over time, a peak disappearing in the Raman spectra over time, the intensity of a peak increasing in the Raman spectra over time, the intensity of a peak decreasing in the Raman spectra over time, sharpness of a peak changing in the Raman spectra over time, the wavenumber of a peak shifting in the Raman spectra over time, and combinations thereof. Said peak or peaks may preferably comprise one or more peaks described in statements of invention for the first aspect of the invention above.

[0092] According to a fourth aspect of the invention, there is provided the use of Raman spectroscopy to predict bone-building drug- or bone-building therapy-induced reversal of bone damage in a subject.

[0093] Statements of invention for the previous aspects of the invention may also be applied mutatis mutandis to the fourth aspect of the invention.

[0094] Detailed Description of the Invention

[0095] In order that the invention may be more clearly understood, embodiments thereof will now be described, by way of example only, with reference to the accompanying drawings, of which:

[0096] Figure 1 shows overlaid subtraction Raman spectra (showing Raman intensity versus wavenumber (cm1)) for comparison between mean treated (48 weeks) and untreated (96 weeks) donors, as described in the study below. N = 38 (untreated week 96) and 95 (treated week 48). Bold lines = treated-untreated*0.95. Faded lines = untreated-treated*0.95.

[0097] Figure 2 shows the results of the partial least squares (PLS) regression analysis performed on spectral data obtained from the study described below. Figures 2a and 2b shows histograms of the proportion of each group against the regression result band for (2a) training and (2b) test sets. Grey shading indicates overlap of the black (treatment) and white (naive) fills. Figure 2c shows regression coefficients used in the PLS regression model versus wavenumber (cm-1). Figure 3 shows % change in treatment score versus interval from baseline / weeks for follow-up nail samples taken for treated and control samples, as described in the study below.

[0098] Summary of study performed

[0099] Raman spectroscopic analysis of fingernail tissue was performed. The tissue was sourced from bisphosphonate treated and untreated patients who were diagnosed as osteoporotic.

[0100] Study design

[0101] Fingernail samples were collected as part of a study which comprised a 2 year, open label, parallel, randomised control intervention trial of three orally administered bisphosphonates. Healthy premenopausal women were also recruited for a parallel control and reference group.

[0102] The study was approved by the Sheffield Research Ethics Committee and the Medicines and Healthcare Products Regulatory Agency and was carried out in accordance with the Declaration of Helsinki and the International Conference on Harmonisation Good Clinical Practice guidelines. Written informed consent was obtained for all individual participants included in the study.

[0103] Study population

[0104] Postmenopausal women were recruited with osteoporosis defined by dual energy X-ray absorptiometry BMD at the lumbar spine or proximal femur of (i) T-score < -2.5 or (ii) T-score < -1.0 plus a prevalent nontraumatic fracture. The participants were recruited through a hospital metabolic bone clinic and from general practice registers. Inclusion criteria were that the women were ambulatory, less than 85 years old, more than 5 years postmenopausal, and able to give informed consent.

[0105] Inclusion criteria

[0106] The group of healthy premenopausal women included ages 35 to 40 years, undergoing regular menstrual cycles and non-use of hormonal contraception.

[0107] Exclusion criteria

[0108] For both groups, these were the presence of fractures in the previous 12 months, the use of medications or diagnosis of any disease or medical condition known to affect bone, or a body mass index (BMI) outside the range of 18 to 35 kg / m2.

[0109] Study interventions

[0110] The women diagnosed as osteoporotic were randomised to receive one of three oral bisphosphonates (open label) at the licensed dose: (i) Ibandronate (Bonviva, Roche; 150 mg once a month), (ii) Alendronate (Fosamax, Merck; 70 mg once a week), or (iii) Risedronate (Actonel, Wamer-Chilcott; 35 mg once a week). To minimise bias, the bisphosphonates were prescribed under a coding system using a stratified block randomisation method. Adherence was assessed using the medical events monitoring system bottle caps (AARDEX, Zurich, Switzerland). In keeping with usual clinical practice, participants also received 3 g calcium carbonate (1200 mg elemental calcium) and cholecalciferol 20 mg (800 IU) per day (Adcal D3, two tablets daily, ProStrakan) which was initiated 1 week before the bisphosphonate at the baseline visit. The healthy premenopausal women were not prescribed any medications throughout the study.

[0111] Nail collection Fingernail clippings from one fingernail on at least one hand were taken from each patient with scissors. A minimum sample depth of 2 mm was collected. The sample was placed into an Eppendorf tube and stored at ambient temperature and humidity.

[0112] Sample analysis using Raman spectroscopy

[0113] The fingernail samples were analysed using Raman spectroscopy. Briefly, a Sierra Reader (Snowy Range Instruments, Laramie, WY, USA) was used at 785 nm excitation with 50 mW power at the sample. The nails were inspected visually to confirm they were free of contamination, and then the nails were placed so that their upper surface faced the laser exit aperture. No further sample preparation was undertaken. The Raman data collected from the nails were processed in Matlab 2013a (Natick, MA, USA) using models created in a previous study (M. R. Towler el al, Clin. Med. Insights Arthritis Musculoskelet. Disord. 2016, 9, 109). A principal component analysis (PCA)-based background removal model derived from an archive of 1500 nails was applied to the data which was then normalised to the first PC score. Spectra were acquired from 400 to 1800 cm-1.

[0114] Statistical analysis

[0115] A total of 460 nails were measured and spectral data from matched nails (different hands from the same donor on same week) were averaged prior to further analysis giving 344 donor-visit spectra. Three spectra did not pass the quality control checks used in the Osentia software, which checks for saturation, low signal to noise, and match to the expected nail spectrum, and so were excluded from further analysis. Spectral data from 282 donor-visits (81 treatment and 201 treatment naive) were used for training the partial least squares (PLS) regression using k = 5 cross validation, with 59 (19 treatment and 40 treatment naive) donor-visits used to validate the model in a held-out test set. Set membership of the spectra was determined at donor level, so any multiple week samples for an individual donor were placed in the same set and no donors contributed to each set. Three components were selected for the model based on the minimum standard error of cross validation. The performance of the PLS regression model was assessed using area under the curve (AUC) for the receiver operator characteristics, with the standard error and confidence intervals (95%) calculated as per Hanley and McNeil (J. A. Hanley, B . J. McNeil, 1, https: / / doi.Org / 10.1148 / radiology.148.3.6878708 1983, 148, 839).

[0116] Spectral changes from baseline A number of individuals were resampled after intervals after their baseline visit, after 12 and 48 weeks for treated groups and after 96 weeks for treatment naive. The percent change in the treatment model score between baseline and each visit was calculated for each individual. The mean change and its 95% confidence interval were calculated for each visit. Raman spectra comparing the differences between mean treated (48 weeks) and untreated (96 weeks) donors are displayed in Figure 1.

[0117] Table 1 below displays the assignment of common bands in the Raman spectra of proteins.

[0118] Table 1

[0119] Visual inspection of Figure 1 in the subtractions of the mean spectra at the final followup for each group, compared with assignments in Table 1 above, shows that the controls exhibit a-helix related bands at 928, 1283, and 1653 cm1and bands related to P-sheet at 1228 and 1674 cm1. Treatments show a-helix-related bands at 935, 1295, and 1653 cm1and P-sheet-related bands at 995 and 1680 cm1. The bands at positions typically assigned to less systematic (‘random’) secondary structure are less consistent, with the range 1660-1670 cm1being approximately equal, the range 1245-1255 cm'1higher in treatment, and the range 1280-1290 cm'1higher in controls. PLS regression versus treatment

[0120] The regression trained against treatment status achieved an optimal standard error of cross-validation (SECV) using 3 pls components, which gave an AUC of 0.77 (95% confidence interval = 0.72-0.83) for the training set, while the test set achieved an AUC of 0.71 (95% confidence interval = 0.57-0.86). Figure 2 shows how the different groups are distributed with respect to the regression score. In both the training and test sets, the treatments have a distribution shifted to a higher value indicating a stable model.

[0121] The regression coefficients (Figure 2c) for the treatment model are complex, without very strong dominant subtraction features; however, numerous bands observed in Figure 1 are again evident in the coefficients, as marked on Figure 2c.

[0122] Within-subject versus between-group variation

[0123] Baseline and follow-up nail samples were taken for 38 treated samples (0, 12, and 48 weeks) and nine untreated donors (0 and 96 weeks). By comparing within-subject changes from baseline to follow-up, any influence due to age is standardised (Figure 3) and this shows there is no overlap in the lower 95% confidence interval at 48 weeks (37% to 377%) in treated donors and the upper 95% confidence interval for either treated 12 week donors (-198% to 34%) or untreated 96 week donors (-92% to -674%). Interestingly, the decrease in treatment score at 48 weeks (207%) is comparable in magnitude to half the increase (383%) observed over 96 weeks in the untreated group.

[0124] Discussion

[0125] The data obtained suggests that, in the absence of intervention, over time, there is a progressive rearrangement of the secondary protein structure of the keratinized tissue in the controls with the accumulation of nonfibrous structures as the proteins reorganise. It is noteworthy that bands assigned to disulphide vibrational modes also play a role in the spectroscopic changes observed as disulphide bonds are critical in imposing and maintaining a protein's higher order structure and their ability to form is affected by local changes in secondary structure. Within the human nails analysed in the current study, the spectral region associated with the disulphide bridges indicates substantial rearrangement of the protein's tertiary and / or quaternary structure. The overall picture of protein structure is therefore that in the absence of intervention, the proteins are becoming less fibrous and less cooperative. The administration of the bisphosphonates reverts this gradual degradation, returning to a more fibrous and cooperative protein. This suggests that the bisphosphonates are providing a mechanism for the tissues to repair the protein rearrangements.

[0126] The model is exploiting specific changes that are associated with the secondary and tertiary structures of keratin. Data previously obtained by the inventors in a rat model of osteoporosis and intervention by pharmacological treatments demonstrated that keratinized tissue was a direct surrogate of changes in bone protein through parallel measurements performed in bone and in keratinized tissue (M. Towler et al, BBA - Molec. Basis Dis. 2018, 1864, 398 and M. Towler et al, J. Mater. Sci. Mater. Med. 2019, 30, 25). However, the present invention surprisingly demonstrates that Raman spectroscopy can be used to successfully predict active reversion of bone damage as a result of bone-building drug / therapy-based treatment, which has not been shown in any of the previous studies performed.

[0127] Because regression coefficients are standardised effects (the data reduction is based on within-group versus between-group variation), any differences observed in the subtraction spectra but not the regression coefficients are therefore features that are too variable to be of diagnostic value. It is interesting to note that the feature at 623 cm1, attributable to free cysteine S-H vibrations, is strong in the subtraction spectrum but absent entirely from the regression coefficients. In order for the protein restructuring to occur, it is essential to break some S-S bonds to form 2x S-H and then reorient the protein and subsequently reform the new S-S linkages in a different conformer. From this scenario, a high flux in S-S linkages and free S-H bonds would be expected, meaning a very high variability within each group, which would then overwhelm any intergroup difference, leading to the feature disappearing from the regression model. A key pathway that bisphosphonates influence bone resorption is through inhibition of the cysteine protease, cathepsin K, which cleaves proteins adjacent to sulfhydryl groups. A serious side effect of high-dose bisphosphonate use (typical for preventing bone metastasis in cancer patients), necrosis of bone tissue, is associated with elevated osteoclast count and expression of cathepsin K to counter the inhibition. In fact, cathepsin K is a focal connection between keratin and collagen, with cathepsin K requiring interaction with a glycosaminoglycan form of keratin (keratin sulphate) in order to be able to lyse the keratin. Cathepsins are involved in the function of keratinocytes responsible for forming keratin-based tissues, and bisphosphonates have variable inhibition selectivity for specific cathepsins. Even bisphosphonates developed to promote selectivity to cathepsin K exhibit activity against cysteine protease cathepsins such as S and L.

[0128] Overall, this study demonstrates that Raman spectroscopic measurement of human keratin can detect a measurable impact on protein order as a result of bone-building drug / therapy administration. Spectroscopic analysis can be used to monitor subtle changes in keratin protein structure, resulting from reversal in bone damage, thus providing an accurate prediction of drug / therapy-induced reversal of bone damage.

[0129] Further, the study shows that measurable changes are observable in spectra taken at even very early stages of treatment (analysed data was from only 48 weeks into treatment). This demonstrates that Raman spectroscopy can be used to detect therapeutic reversion of bone damage soon after the onset of treatment, in comparison to other known methods which are only capable of detecting reversion of bone damage much longer after the onset of treatment (e.g. at least 2 years for Dual-energy X-ray absorptiometry (DXA)).

[0130] The above embodiments are described by way of example only. Many variations are possible without departing from the scope of the invention as defined in the appended claims.

Claims

CLAIMS1. A method for predicting the extent of bone-building drug- or bone-building therapy-induced reversal of bone damage in a subject who has been administered at least one bone-building drug and / or has undergone bonebuilding therapy, the method comprising the steps of: a. Obtaining a keratinized tissue sample from the subject; b. Obtaining a Raman spectrum of the keratinized tissue sample; c. Analysing the Raman spectrum to identify at least one marker indicative of bone-building drug- or therapy-induced reversal of bone damage; and d. Predicting the extent of bone-building drug- or therapy-induced reversal of bone damage based on the identification of the at least one marker in the Raman spectrum.

2. A method as claimed in claim 1, wherein the at least one bone-building drug comprises at least one bisphosphonate.

3. A method as claimed in any preceding claim, wherein the subject is a human.

4. A method as claimed in any preceding claim, wherein the subject has a bone- related disease or condition that is independently selected from the group consisting of: a bone fracture, osteoporosis, osteoarthritis, rheumatoid arthritis, Paget’s disease, osteogenesis imperfecta, osteomalacia, scoliosis, kyphosis, lordosis, bone cancer, fibrous dysplasia, osteochondritis dissecans, avascular necrosis, osteomyelitis, and combinations thereof.

5. A method as claimed in any preceding claim, wherein the keratinized tissue sample is independently selected from the group consisting of: a nail sample, a hair sample, a skin sample, and combinations thereof, and wherein the keratinized tissue sample preferably comprises a nail sample.

6. A method as claimed in any preceding claim, wherein step (b) comprises obtaining the Raman spectrum of the keratinized tissue sample in situ and when the sample is attached to the body of the subject.

7. A method as claimed in any one of claims 1 to 5, wherein step (b) comprises obtaining the Raman spectrum of the keratinized tissue sample when it is not attached to the body of the subject.

8. A method as claimed in any preceding claim, wherein step (c) comprises subjecting the Raman spectrum to a multivariate analysis technique to identify the at least one marker indicative of bone -building drug- or therapy-induced reversal of bone damage, wherein the multivariate analysis technique preferably comprises partial least squares (PLS) regression.

9. A method as claimed in any preceding claim, wherein step (c) comprises the step of comparing the obtained Raman spectrum with at least one reference spectrum or reference spectral data to identify the at least one marker indicative of bone-building drug- or therapy-induced reversal of bone damage.

10. A method as claimed in claim 9, wherein the at least one marker comprises a change relative to the at least one reference spectrum or reference spectral data in one or more peaks in the obtained Raman spectrum of the keratinized tissue sample.

11. A method as claimed in claim 10, wherein the change in the one or more peaks is independently selected from the group consisting of: a newfound presence of the one or more peaks in the obtained spectrum compared to the at least one reference spectrum or reference spectral data, disappearance of the one or more peaks in the obtained spectrum compared to the at least one reference spectrum or reference spectral data, increase in intensity of the one or more peaks in the obtained spectrum compared to the at least one reference spectrum or reference spectral data, decrease in intensity of the one or more peaks in the obtained spectrum compared to the at least one reference spectrum or reference spectral data, change in the sharpness of the one or more peaks in the obtained spectrum compared to the at least one reference spectrum or reference spectral data, a shift in the wavenumber of the one or more peaks in the obtained spectrum compared to the at least one reference spectrum or reference spectral data, and combinations thereof12. A method as claimed in claim 10 or 11, wherein the at least one marker comprises a change relative to the at least one reference spectrum or reference spectral data in one or more peaks in the obtained Raman spectrum of the keratinized tissue sample that denote the level of sulfur bonding in the keratinized tissue sample.

13. A method as claimed in claim 12, wherein the at least one marker comprises a change relative to the at least one reference spectrum or reference spectral data in one or more peaks in the obtained Raman spectrum of the keratinized tissue sample that denote cysteine-cysteine S-S bond stretching, and wherein said one or more peaks are optionally independently selected from the groupconsisting of: the gauche-gauche-gauche (GGG) conformation of cysteinecysteine S-S bond stretching, the gauche-gauche-trans (GGT) conformation of cysteine-cysteine S-S bond stretching, and combinations thereof.

14. A method as claimed in any one of claims 10 to 13, wherein the at least one marker comprises a change relative to the at least one reference spectrum or reference spectral data in one or more peaks in the obtained Raman spectrum of the keratinized tissue sample that denote an a-helix and / or P-sheet secondary structure in the keratinized tissue sample.

15. A method as claimed in claim 14, wherein the at least one marker comprises a change relative to the at least one reference spectrum or reference spectral data in one or more peaks in the obtained Raman spectrum of the keratinized tissue sample that denote C-C stretching in a P-sheet secondary structure.

16. A method as claimed in claim 14 or 15, wherein the at least one marker comprises a change relative to the at least one reference spectrum or reference spectral data in one or more peaks in the obtained Raman spectrum of the keratinized tissue sample that are found within the amide III spectral region and denote an a-helix secondary structure, optionally a globular or fibrous a- helix secondary structure.

17. A method as claimed in any one of claims 14 to 16, wherein the at least one marker comprises a change relative to the at least one reference spectrum or reference spectral data in one or more peaks in the obtained Raman spectrum of the keratinized tissue sample that are found within the amide I spectral region and denote an a-helix secondary structure.

18. A method as claimed in any one of claims 14 to 17, wherein the at least one marker comprises a change relative to the at least one reference spectrum or reference spectral data in one or more peaks in the obtained Raman spectrum of the keratinized tissue sample that are found within the amide I spectral region and denote a P-sheet secondary structure.

19. A method as claimed in any one of claims 10 to 18, wherein the at least one marker comprises a change relative to the at least one reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 400-700 cm1, or between 500-600 cm1.

20. A method as claimed in any one of claims 10 to 19, wherein the at least one marker comprises a change relative to the at least one reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 970-1010 cm1, or between 990-1005 cm1.

21. A method as claimed in any one of claims 10 to 20, wherein the at least one marker comprises a change relative to the at least one reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 1200-1350 cm1, or between 1220-1320 cm1, or between 1265-1320 cm-1.

22. A method as claimed in any one of claims 10 to 21, wherein the at least one marker comprises a change relative to the at least one reference spectrum or reference spectral data in one or more peaks in the Raman spectral range of between 1645-1680 cm1.

23. A method of monitoring bone-building drug- or bone-building therapy- induced reversal of bone damage over time in a subject who has been administered at least one bone-building drug and / or has undergone bonebuilding therapy, the method comprising the steps of: a. Obtaining multiple keratinized tissue samples from the subject over a period of time; b. Obtaining a Raman spectrum of each obtained keratinized tissue sample; c. Analysing each Raman spectrum to identify at least one marker indicative of bone-building drug- or therapy-induced reversal of bone damage; and d. Monitoring changes in the at least one identified marker over time to assess the progression of bone-building drug- or therapy-induced reversal of bone damage.

24. Use of Raman spectroscopy to predict bone-building drug- or bone -building therapy-induced reversal of bone damage in a subject.

25. A method for predicting the extent of bone-building drug- or bone-building therapy-induced reversal of bone damage in a subject who has been administered at least one bone-building drug and / or has undergone bonebuilding therapy, the method comprising the steps of: a. Obtaining a keratinized tissue sample from the subject; b. Obtaining a Raman spectrum of the keratinized tissue sample;c. Comparing the obtained Raman spectrum of the keratinized tissue sample with at least one reference spectrum or with reference spectral data; and d. Predicting the extent of bone-building drug- or therapy-induced reversal of bone damage based on the results of the comparison of the obtained Raman spectrum with the at least one reference spectrum or with the reference spectral data.

Images