Patent Application
20160077051
The present invention relates to a method of determining tumour response to chemotherapy by comparing the RNA integrity of tumour cells before, during and after chemotherapy.
Cancer is the uncontrolled malignant growth of cells. In a process called metastasis, cancerous cells can spread from their site of origin to distant sites within the body, via the lymphatic and/or circulatory systems. Metastasis is the leading cause of death in humans with cancer (Bockhorn, M. et al., Lancet Oncol. 8 (2007) 444-448.)
There are a number of treatments which are used to treat or control cancer, including surgery, radiation therapy and chemotherapy. Surgery and radiation therapy are typically used to remove non-metastatic cancerous tumours (abnormal growths composed of cancerous cells). However, the presence of metastatic cancer necessitates the use of systemic chemotherapy regimens to combat the growth of primary tumours (before or after surgery) and secondary tumours throughout the body. In breast cancer, effective systemic chemotherapy agents include the anthracyclines (typically doxorubicin or epiuribicin), taxanes (paclitaxel or docetaxel), nucleoside analogs (5-fluorouracil), and alkylating agents (cyclophosphamide) (Parissenti, A. M. et al. Anticancer Drugs 18 (2007) 499-523). Anthracyclines disrupt the uncoiling of DNA by topoisomerase II (“topo II”), intercalate between DNA strands, and cause DNA lesions, thereby interfering with DNA replication in rapidly dividing tumour cells. Taxanes, on the other hand, block the depolymerization of microtubules, resulting in an arrest of the cell cycle at mitosis and the subsequent induction of apoptosis (Distefano, M. et al., Int. J. Cancer 72 (1997) 844-850; Moos, P. J. et al., Proc. Natl. Acad. Sci. U.S.A 95 (1998) 3896-3901). The nucleoside analog 5-fluorouracil blocks the conversion of dUMP into dTMP, while the alkylating agent cyclophosphamide forms covalent bonds with DNA (Parker, W. B. et al., Pharmacol Ther. 48 (1990) 381-395; Bignold, L. P., Anticancer Res. 26 (2006) 1327-1336). These latter two drugs disrupt DNA replication in rapidly dividing cells at S phase (Zijlstra, J. G. et al. Oncol. Tumor Pharmacother. 7 (1990) 11-18; Richardson, D. S. et al., Blood Rev. 11 (1997) 201-223; Capranico, G. et al., Chem. Biol. Interact. 72 (1989) 113-123; Chazard, M. et al., Bull. Cancer 81 (1994) 173-181).
Most drugs that are used in chemotherapy are highly cytotoxic, and destroy both healthy normal cells (particularly if they are rapidly dividing) and cancerous cells. As such, chemotherapy drugs cause significant side effects, such as immunosuppression, nausea and vomiting, and cardiotoxicity. These side effects can have a significant negative effect on the patient's quality of life.
Tumour response to chemotherapy agents can vary widely between patients, due to the presence of drug resistance mechanisms in some patients that block drug efficacy. Drug resistance can be “intrinsic” (i.e. pre-exist in the tumour) or “acquired” through continued exposure to chemotherapy agents. A number of mechanisms have been identified which play a role in reduced responsiveness of tumour cells to chemotherapy agents in vitro. For the anthracyclines and taxanes, these include the overexpression of drug transporters (e.g. P-glycoprotein) and the multidrug resistance protein, the downregulation of topoisomerase II α, mutations in the cell cycle regulator protein p53, the increased synthesis of thymidylate synthase or the drug-conjugating enzyme glutathione-S-transferase, and the accumulation of mutations in genes coding for the α or β chains of tubulin (Juliano, R. L. et al., Biochim. Biophys. Acta 455 (1976) 152-162; Beck, W. T. et al. Cancer Res. 39 (1979) 2070-2076; Cole, S. P. et al., Science 258 (1992) 1650-1654; Fry, A. M. et al., Cancer Res. 51 (1991) 6592-6595; Giaccone, G. et al. Cancer Res. 52 (1992) 1666-1674; Balcer-Kubiczek, E. K. et al. Radiat. Res. 142 (1995) 256-262; Aas, T. et al., Nat. Med. 2 (1996) 811-814; Batist, G. et al., J. Biol. Chem. 261 (1986) 15544-15549; Batist, G. et al., Biochem. Pharmacol. 35 (1986) 2257-2259; Harris, A. L. et al., Acta Oncol. 31 (1992) 205-213; Cabral, F. et al., Proc. Natl. Acad. Sci. U.S.A 78 (1981) 4388-4391; Schibler, M. J. et al., J. Cell Biol. 113 (1991) 605-614).
Recently, genome profiling approaches have provided significant insight into the genes and mechanisms associated with the acquisition of drug resistance in breast tumour cells. (Parissenti, A. M. et al., Anticancer Drugs 18 (2007) 499-523; Villeneuve, D. J. et al., Breast Cancer Res. Treat. 96 (2006) 17-39).
The presence of multiple and varied mechanisms of intrinsic or acquired drug resistance makes it very difficult to identify which patients will respond to a given chemotherapy regimen and whether this response will be sustained throughout treatment. In patients, a sensitive tumour may regress or shrink during chemotherapy, and continue to regress following chemotherapy. In other patients, a resistant tumour can be unresponsive to chemotherapy both mid- and post-treatment. Finally, a tumour may regress during chemotherapy in some patients, but return to its original state (or continue to grow) after chemotherapy is completed.
It would be highly beneficial to be able to determine the level of tumour responsiveness to a given chemotherapeutic drug or agent before administration, or early after drug administration. For example, only 33% and 35.4% of breast cancer patients respond to paclitaxel and docetaxel after anthracycline-based chemotherapy, respectively (Seidman, A. D. et al., J. Clin. Oncol. 13 (1995) 1152-1159; Ando, M. et al., J. Clin. Oncol. 19 (2001) 336-342). However, biomarkers capable of distinguishing between chemotherapy-sensitive and chemotherapy-resistant tumours in cancer patients have yet to be identified. Thus, for cancer patients receiving chemotherapy regimens involving cytotoxic agents, there is no current method to determine whether a tumour is responding to chemotherapy mid-treatment or whether the viability of tumour(s) has been eradicated post-treatment. Consequently, cancer patients experience the serious negative side effects from taking cytotoxic drugs, without knowing whether their tumours are, in fact, responding to these agents.
Accordingly, there is a need for a method of quickly and accurately assessing the level of responsiveness of tumours to particular chemotherapy drugs (and combinations thereof), in order to tailor a specific regimen best suited to a patient's needs. There is a further need for indicators of sensitivity or resistance to chemotherapy drugs.
In accordance with a broad aspect of the present invention, there is provided a method of determining tumour responsiveness to one or more chemotherapeutic agent(s) in a patient with one or more cancerous tumours, comprising:
wherein a decrease in the RNA quality over said period of time indicates that the tumour is responsive to the chemotherapeutic agent(s).
In an embodiment of the invention, the RNA quality is determined as a ratio of 28S and 18S rRNA intensity values, wherein said ratio is obtained by gel electrophoresis of the extracted RNA, ethidium bromide staining of said gel, and calculation of said ratio of intensities of 28S and 18S rRNA visualized under ultraviolet light.
In another embodiment of the invention, the RNA quality is determined by capillary electrophoresis of the extracted RNA and quantification of the various RNAs separated in the electrophoresis. Preferably, the RNA quality is quantified as an RNA integrity number (RIN), wherein the RIN is calculated by an algorithmic assessment of the amounts of various RNAs present within the extracted RNA.
More than one chemotherapeutic agent may be administered to the patient. The chemotherapeutic agent can be selected from the group consisting of anthracycline and taxane chemotherapeutic agents. Preferably, the chemotherapeutic agents comprise an anthracycline and a taxane. In an embodiment of the invention, epirubicin and docetaxel are used in the chemotherapy regimen.
In yet another preferred embodiment, the RNA is extracted from one or more core biopsies of a tumour of the patient. Preferably, the core biopsy is obtained by image-guided means such as computed tomography (CT), x-ray, ultrasound, and magnetic resonance imaging (MRI). The RNA quality is then determined from the one or more core biopsies of said tumour.
In another embodiment of the invention, the magnitude of reduction in the RNA quality is proportionate to tumour responsiveness, wherein tumour responsiveness may be assessed by a corresponding decrease in tumour extent and/or cellularity, and clinical response of the patient.
In another embodiment of the invention, a patient with a post-treatment RIN of 3 or less is identified as being responsive to the chemotherapeutic agent(s), and a patient with a post-treatment RIN of 3 or more is identified as being non-responsive to the chemotherapeutic agent(s).
In yet another aspect of the invention, there is provided a use of tumour RNA quality to determine a patient's responsiveness to one or more chemotherapeutic agents, wherein the RNA quality of tumour cells is determined before administration of the chemotherapeutic agent(s), and compared with the RNA quality of tumour cells after administration of the one or more chemotherapeutic agents, and a decrease in the RNA quality after administration of the one or more chemotherapeutic agents indicates that the patient is responsive to the chemotherapeutic agent(s). In a preferred embodiment, tumour RNA quality is quantified as an RNA integrity value (RIN). In this embodiment, the use comprises determining a first RIN of tumour cells obtained from the patient before administration of the chemotherapeutic agent, and comparing the first RIN to the RIN of tumour cells determined during and/or after administration of the one or more chemotherapeutic agents, wherein a decrease in the RIN during and/or after administration of the one or more chemotherapeutic agents indicates that the patient is responsive to the one or more chemotherapeutic agents.
An advantage of the present invention is that tumour RNA quality, quantified as an RNA integrity number (RIN), is an easily accessed biomarker of tumour responsiveness to a particular chemotherapy regimen involving one or more chemotherapeutic agents.
Presently known methods of determining tumour responsiveness, which generally require the visual interpretation of photomicrographs of fixed and stained sections of core biopsies by a human operator such as a pathologist. Such methods are dependent on the subjective interpretation by the operator, which may vary from one person to the next. Such methods are also prone to human error. The assessment of tumour RIN provides a significant advantage over presently known methods of assessing tumour responsiveness to a given chemotherapy regiment, as the tumour RIN is a quantitative biomarker of tumour responsiveness that is both accurate and reproducible.
Another advantage of the present invention is that assessment of tumour RIN can be carried out by automated means. The automated means can involve high-throughput screening, which allows for rapid assessment of tumour RIN. The rapid assessment of tumour RIN thus allows rapid and accurate assessment of the level of responsiveness of a patient's tumour to a given chemotherapy regimen.
Another advantage of the present invention is that the RIN value of tumour cells may be correlated with the dosage level of the chemotherapeutic agent, thus allowing tailoring of a chemotherapy regimen to a patient's needs, or level of responsiveness.
Other and further advantages and features of the invention will be apparent to those skilled in the art from the following detailed description of an embodiment thereof, taken in conjunction with the accompanying drawings.
The present invention will be further understood from the following detailed description of an embodiment of the invention, with reference to the drawings in which:
Ribonucleic acids (RNAs) play a number of essential roles in the translation of genetic information into functional proteins within eucharyotic cells. mRNAs are processed transcripts of genes, which bind to ribosomes for translation into specific proteins. Other RNAs form vital portions of ribosomes (rRNAs), or act as carriers for amino acids in protein synthesis (tRNAs).
The levels of cellular RNA are precisely regulated, maintaining a balance between transcription and RNA degradation pathways. There is increasing evidence that surveillance systems are present in cells to monitor RNA quality within the cell. These surveillance systems are coupled to RNA degradation pathways to rid cells of a variety of defective RNAs (Houseley, J. et al., Nat. Rev. Mol. Cell Biol. 7 (2006) 529-539; Parker, R. et al., Nat. Struct. Mol. Biol. 11 (2004) 121-127). Defective RNAs include inappropriately processed primary transcripts and mRNAs lacking translational stop codons, containing premature termination codons, or containing nonsense codons. In addition, there is evidence that RNA degradation is a hallmark of apoptosis (programmed cell death). Apoptosis-inducing agents have been shown to induce RNA degradation in cells (King, K. L. et al, Cell Death. Differ. 7 (2000) 994-1001; Bakhanashvili, M. et al., J. Mol. Med. (2007)). In particular, chemotherapy agents, particularly those generating reactive oxygen species, may induce sufficient damage to DNA and/or RNA, such that a variety of defective RNAs are produced and the above-noted RNA degradation pathways are activated.
Given the critical role which the various cellular RNAs play in cell function, the types of RNA and their intracellular concentrations can provide a significant amount of information on cellular activity, such as gene expression and protein production. Thus, it is desirable to extract cellular RNA in order to obtain a “snapshot” of what is happening within the cell at a given point in time. The extracted RNA can then be used to clone cDNAs into expression vectors, to identify and quantify mRNA transcripts by reverse transcription polymerase chain reaction (RT-PCR), and gene expression profiling by high-throughput RT-PCR or microarray studies. However, since RNA is susceptible to extensive degradation by RNAse enzymes which are ubiquitous in the environment, an assessment of RNA quality or integrity is essential before performing the above applications. The RNA “quality” or “integrity” (used interchangeably throughout) thus refers to the state of the RNA following extraction from the cell. High RNA quality is taken as meaning little to no degradation of the RNA following extraction, whereas low RNA quality means the extracted RNA exhibits a significant to total degradation.
In the past, RNA quality or integrity has been evaluated by visualization of RNA bands under ultraviolet light after gel electrophoresis and staining of the gel with ethidium bromide. Typically, the intensity values for the 28S and 18S rRNA bands are determined by film densitometry and a 28S/18S rRNA ratio computed. RNA is considered of high quality if the 28S/18S rRNA ratio is about 2.0 or higher. However, since the above approach relies on the interpretation of the gel and/or film densitometry by a human operator, it is subjective and the results are difficult to reproduce between different operators. In addition, large quantities of RNA are also required for this approach, making it difficult to obtain enough RNA for an analysis.
Recently, microcapillary electrophoresis has been used increasingly to assess RNA integrity, particularly since only nanogram quantities of RNA are required. One such platform, the Agilent® 2100 Bioanalyzer (Agilent Technologies, Inc., U.S.A.) uses microfluidics technology to carry out electrophoretic separations of RNAs in an automated, reproducible manner (Mueller, O. et al., Electrophoresis 21 (2000) 128-134). The Agilent® 2100 Bioanalyzer is now used in many laboratories for the assessment of RNA quality, particularly following the development of software for the Agilent® Bioanalyzer that calculates an RNA integrity number (RIN) for each sample after capillary electrophoresis. (Schroeder, A. et al., BMC. Mol. Biol. 7 (2006) 3; Imbeaud, S. et al. Nucl. Acids Res. (2005), 33, 6, e56, 1-12). This software incorporates an algorithm which quantifies the amounts of multiple RNAs in the electropherogram of a given RNA sample and assigns a RIN value based on this assessment. Recent studies suggest that the RIN is superior to the 28S/18S rRNA ratio for reliably measuring RNA quality. (Schroeder, A. et al., BMC. Mol. Biol. 7 (2006) 3; Weis, S. et al., J. Neurosci. Methods 165 (2007) 198-209; Strand, C. et al., BMC. Mol. Biol. 8 (2007) 38). The RIN is emerging as the best method for RNA quality assessment in mammalian cell lines and tissues, including tumours of the breast and other organs. (Fleige, S. et al., Biotechnol. Lett. 28 (2006) 1601-1613; Strand, C. et al., BMC. Mol. Biol. 8 (2007) 38).
It has now been discovered that RNA quality or integrity, as measured by tumour RNA quality, particularly as measured with the RIN value, can be used as a direct measure of cell viability.
A set of tumour biopsies were taken from 50 patients with breast cancer, who were undergoing an epirubicin/docetaxel chemotherapy regimen with pegflgrastim support (see Example 2). Two tumour biopsies were taken from each patient at three time points, before (pre-), during (mid-) and after (post-) chemotherapy treatment, to form two sets of biopsies for each patient (each set composed of a pre-, a mid- and a post-treatment biopsy). One set of biopsies was analyzed for the RNA quality (i.e. determination of RIN value) and the other set of biopsies was subjected to immunohistochemical analysis to determine levels of specific tumour marker proteins known to be important for breast cancer prognosis, percentage tumour cellularity and photomicrographs. Tumour RIN was then compared to the observed changes in tumour marker proteins (Example 2(d)), tumour cellularity (Example 2(f)) and photomicrographs (Example 2(g)) that occurred during the course of the chemotherapy treatment, and analyzed for statistically significant correlation between tumour RIN and the observed changes. Finally, tumour RIN was compared with the observed clinical response of the patient.
Dramatic reductions in RNA integrity of tumour cells were observed to occur in drug-sensitive tumours post-chemotherapy, while drug-resistant tumours were observed to retain high RNA integrity, resulting in disease progression and poor patient prognosis. As noted in Example 2(b), a high drug dose level was strongly associated with a large negative change in tumour RIN during the course of treatment. Also, a low drug dose level correlated with few or small reductions in tumour RIN. This suggested that a reduction in tumour RIN was directly related to chemotherapy drug response in these patients. A strong positive correlation as found to between tumour extent (cellularity) and tumour RIN values measure post-treatment (see Example 2(e)). That is, a decrease in tumour extent was proportionate to the decrease in tumour RIN. Finally, tumour RIN measured post-treatment was found to be an accurate predictor of tumour response to chemotherapy and observed clinical response (see Example 2(f),(g)).
The response of tumours to a specific chemotherapy regimen in cancer patients can thus be effectively determined by monitoring the ability of the regimen to induce a reduction in tumour RNA quality (integrity).
Tumour RNA integrity can be measured by capillary electrophoresis, followed by the assignment of a RIN value. Chemotherapy-induced reductions in tumour RIN values would be indicative of responsive tumours, while little change in tumour RIN values would suggest that the tumour is resistant to the selected regimen.
In order to determine a cancer patient's responsiveness to a chemotherapy regimen, RNA is extracted from the patient's tumour(s) at least two different time points during the administration of a chemotherapy regimen. Preferably, RNA is extracted from the tumour before the administration of a chemotherapy regimen, and during and/or after completion of the regimen.
The chemotherapy regimen can consist of one chemotherapy agent or a combination of two or more chemotherapy agents, and the doses of each agent may be varied with time.
To improve reproducibility and accuracy, the tumour cells are preferably collected in one or more image-guided biopsies. To further improve reproducibility and accuracy, three or more image-guided biopsies are collected from the tumour. An image-guided biopsy is obtained with image-guided means such as computed tomography (CT), x-ray, ultrasound, and magnetic resonance imaging (MRI).
The quality of the extracted RNA is then determined. This can be done by traditional means such as obtaining the 28S/18S rRNA ratio as noted above. However, RNA quality is preferably determined by capillary electrophoresis of the extracted RNA and quantification of the RNAs in the resultant electropherogram. An automated analytical system, such as the Agilent® 2100 Bioanalyzer (Agilent Technologies, Inc., U.S.A.), is preferred for carrying out this determination, as such a system can assess the electropherogram and quantify the quality of a given RNA sample as an RNA integrity number (RIN). The Agilent® 2100 Bioanalyzer calculates the RIN using an algorithm which is incorporated in the software associated with the Bioanalyzer (Schroeder, A. et al., BMC. Mol. Biol. 7 (2006) 3; Imbeaud, S. et al. Nucl. Acids Res. (2005), 33, 6, e56, 1-12). The resultant RIN values are reproducible from one operator to the next, and can be processed digitally. Moreover, an automated analytical system such as the Agilent® 2100 Bioanalyzer allows rapid, high-throughput analyses of RNA samples. Thus, a patient's responsiveness to a given chemotherapy regimen can be determined during or after a chemotherapy regimen, and the regimen may be changed after no tumour response is detected. This is of great benefit as it identifies patients that have not responded to the chemotherapy regimen and would likely be at high risk of disease progression.
The RIN value of the tumour cells collected before administration of the chemotherapy regimen is then compared with the one or more RIN values of tumour cells collected after commencement of the regimen, i.e. during and/or after completion of the chemotherapy regimen. If a patient exhibits no change in tumour RNA integrity during treatment (response pattern A as noted in Example 2), then the patient's tumour would be considered resistant to the chemotherapy regimen being used. The patient would be considered non-responsive to the chemotherapy regiment, i.e. at high risk of tumour progression and prognosis would be considered poor. Alternative chemotherapy regimens or treatment protocols can then be considered, such as a change in dosage level and/or a change in the type of chemotherapy agent(s) being administered. The method outlined herein can then be repeated to determine responsiveness to the new regimen, thus allowing tailoring of a chemotherapy regimen according to the patient's response.
If a patient exhibits a dramatic reduction in tumour RNA integrity (>50%) both mid- and post-treatment (response pattern C as noted in Example 2), then the patient would be considered to have responded to chemotherapy and would be at lower risk of tumour progression. The patient's prognosis would be considered good. Tumour RIN values near zero would be highly indicative of response to chemotherapy and low risk of tumour progression.
If a patient exhibits a dramatic reduction in tumour RNA integrity post-treatment only (response pattern B as noted in Example 2), then the patient would be considered to have responded to chemotherapy and be at lower risk of tumour progression. The patient's prognosis would be considered good.
If a patient exhibits a dramatic change in tumour RNA integrity mid-treatment only, then she likely has responded to therapy and would be at a lower risk of disease recurrence. This is regardless of a return to high “tumour” RNA integrity post-treatment, since the high quality RNA post-treatment may stem from normal tissue that has infiltrated the lesion. However, it is possible that the tumour has recurred post-treatment.
Further details of the preferred embodiments of the invention are illustrated in the following Examples which are understood to be non-limiting with respect to the appended claims.
(a) Total RNA Isolation from Breast Tissue Core Biopsies.
RNA was isolated from patient tumour core biopsies using QIAGEN® RNAeasy® mini kits (Qiagen GmbH, Germany). The RNA isolation protocol was slightly modified from the protocol published by Qiagen GmbH (freely available from Qiagen GmbH, Germany; also available at http://www1.qiagen.com/literature/handbooks/literature.aspx?id=1000291).
Image-guided needle core biopsies of the patients tumour were taken from the patient, immediately touch prepared to a glass slide for determination of tumour cellularity, and the core biopsy immediately flash frozen on dry ice for future analysis. The frozen core biopsies were immediately dropped in 0.5 ml of RLT buffer containing β-ME (10 μl into 1 ml) in a Eppendorf tube. The biopsies in RLT buffer were homogenized with a Coreless™ motor homogenizer for 5 min (Kontes Glass Company, U.S.A., Cat#:749540-0000).
The lysate was then passaged at least 5 times through a 20-gauge needle (0.9 mm diameter) fitted to an RNase-free syringe. The sample was then centrifuged at high speed in a refrigerated microfuge at 4° C. for 3 min., with transfer of the supernatant to a new tube.
One volume (500 μl) of 70% ethanol was then added to the supernatant and the sample mixed well by repeated pipetting. If some lysate was lost during homogenization, then the volume of ethanol was adjusted accordingly. Visible precipitates formed after the addition of ethanol in some samples did not affect the RNA isolation procedure.
A maximum of 700 μl of the sample, including any precipitate, were added to a RNeasy® mini column and placed in a 2 ml collection tube. The column was centrifuged for 15 s at ˜8000×g (˜10,000 rpm) and the flow-through discarded. The remainder of the sample was then added to the column and the column centrifuged again.
Seven hundred μA of Buffer RW1 was then added to the RNeasy® column and the column centrifuged for 15 s at ˜8000×g (˜10,000 rpm) to wash the column. The flow-through was discarded.
The RNeasy® column was transferred into a new 2 ml collection tube and 500 μl of Buffer RPE was applied to the column. The column was then centrifuged for 15 s at ˜8000×g (˜10,000 rpm) to wash the column. The flow through was discarded.
The RNeasy® column was transferred to a new 2 ml collection tube, discarding the old collection tube and flow-through. The column was then centrifuged again in a microcentrifuge at full speed for 1 min., discarding the collection tube and flow-through once again.
To elute the bound RNA, the RNeasy® column was transferred to a new 1.5 ml collection tube. Thirty μl of RNase-free water was applied directly to the column and the column centrifuged for 1 min. at ˜8000×g (˜10,000 rpm).
To obtain a higher total RNA concentration for the sample, a second elution step was performed using the eluate from step 8.
The concentration and quality of RNA was then checked using an Agilent® 2100 Bioanalyzer and associated software.
The total RNA sample from the tumour core biopsy was applied to RNA 6000 Nano Lapchips™ (Agilent Technologies, Inc.) and subjected to capillary electrophoresis using an Agilent® 2100 Bioanalyzer. The protocol for the Agilent® 2100 Bioanalyzer (Agilent Technologies, Inc.) was followed (Agilent® 2100 Bioanalyzer User's Guide, ed. November 2003, Manual Part No. G2946-90000, Agilent Technologies, Inc., available at http://www.chem.agilent.com/temp/rad4DEAE/00000725.PDF).
The amount of RNA in the sample and the quality of the RNA (RNA integrity) was determined using the RIN algorithm disclosed by Schroeder et al. (Schroeder, A. et al. “The RIN: an RNA integrity number for assigning integrity values to RNA measurements”, BMC. Mol. Biol. 7 (2006) 3.), which is incorporated in the computer software associated with the Agilent® 2100 Bioanalyzer (the software and accompanying manual are freely available from Agilent Technologies Inc., and also at http://www.chem.agilent.com/scripts/generic.asp?lpage=52241 &indcol=N&prodcol=Y)
(a) Tumour Biopsy Samples from Breast Cancer Patients in Chemotherapy Clinical Trial
To test whether treatment of breast cancer patients with chemotherapy agents results in tumour RNA degradation, six image-guided core biopsies of tumours were taken from 50 patients with locally advanced or inflammatory breast cancer pre-, mid-, and post-treatment with epirubicin/docetaxel chemotherapy. Patients were from a national clinical trial hosted by the National Cancer Institute of Canada Clinical Trials Group (referred to as group “MA.22”) and were treated with increasing dose levels of both epirubicin and docetaxel, with pegfilgrastim support to reduce neutropenia associated with this therapy. Chemotherapy was administered in a standard dosing regimen (Arm A) every 3 weeks, and the dose levels used in this study are depicted in Table 1. The maximum tolerated dose for this regimen was dose level 6, i.e. 105 mg/m2 epirubicin and 75 mg/m2 docetaxel (see Schedule A, Table 1).
As noted in Scheme 1, three of the six biopsies taken from each patient were freshly frozen for RNA quality studies, while the remainder were fixed in formalin for assessment of levels of specific tumour markers, including the estrogen receptor ER (Novacastra® Clone 6G11, Leica Microsystems, Germany), the progesterone receptor PR (Novacastra® Clone 16, Leica Microsystems, Germany), topoisomerase II (“Topo II”; clone SWT3D1, Dako Denmark A/S) and human epidermal growth factor receptor 2 (“Her2”; Zymed® TAB250, Invitrogen Corp., U.S.A.). RNA was isolated from two of the freshly frozen core biopsies using RNeasy® Mini kit (Qiagen GmbH, Germany), after which the RNA quality of the sample was assessed by capillary electrophoresis using an Agilent® 2100 Bioanalyzer. The Bioanalyzer quantified the abundance of specific RNAs in the sample and assigned an RNA integrity number (RIN) to each sample. As shown in
The mean core RIN values for RNA isolated from patient core biopsies were then assessed to determine if these values fell in response to epirubicin/docetaxel chemotherapy and whether there was a relationship between drug dose level and the magnitude of RIN reduction. While there was variation in pre-treatment mean tumour RIN values for patients, they were rarely below 5.0, with a mean value of 6.5 when data from all 50 patients was assessed (Table 2). The association between RIN and baseline drug dose was assessed using a 1-way ANOVA.
As expected, there was no association between the magnitude of the tumour RIN and drug dose level at baseline (p=0.45), given that patients had yet to receive chemotherapy. In contrast, RIN values were significantly and negatively correlated with drug dose level mid-treatment (p=0.04). Few or small reductions in the tumour RIN were observed for patients receiving low drug doses, while patients receiving high drug doses exhibited dramatic reductions in the tumour RIN (
0.03
(0.0, 0.0)3
0.03
(0.0, 0.0)3
(c) Patterns of Change in RNA Integrity During Treatment of Patients with Epirubicin/Docetaxel Chemotherapy
Immunohistochemical approaches were then used to determine baseline levels of specific tumour marker proteins known to be important for breast cancer prognosis, and expression was rated as a percentage of positive stain against a known standard. Proteins assessed by immunohistochemistry included the estrogen receptor (ER), the progesterone receptor (PR), human epidermal growth factor receptor 2 (Her2) and topoisomerase II (“topo II”). Associations between pre-treatment levels of specific tumour markers and RIN values at various time points were then assessed by computing Spearman and Pearson correlation coefficients, with or without data transformation to improve symmetry and stabilize data variances. For all patients, high pre-treatment levels of topo II were significantly associated with high tumour RIN pre-treatment (Table 3; p values between 0.01 and 0.03). The association of high pre-treatment RIN values with high pre-treatment topo II levels suggested that cells with high topo II expression are highly viable, rapidly proliferating, and produce high quality RNA. This association was also evident post-treatment (see Table 3). Tumours with high RIN values (i.e. high quality RNA) post-treatment, were taken as representing either highly viable tumours not responding to chemotherapy or healthy normal tissue that had infiltrated the tumour. No association was observed between pre-treatment levels of Her2, ER or PR and RIN values pre- or post-treatment (data not shown).
0.01
0.03
0.02
0.02
0.03
0.04
One measure of drug response in patients involves assessing the magnitude of reduction in the number of tumour cells comprising a lesion(s) post-therapy. As summarized in Table 4 below, the changes in RIN values corresponded to various patterns of change in tumour cellularity values. Overall, patients exhibited significant responses to chemotherapy, given that the overall tumour extent or cellularity fell from 90.94%±2.17% to 50.69%±5.79% mid-treatment and 39.6%±5.78% post-treatment (see Table 4). Patients who exhibited the following response patterns were placed into Groups A to E as follows: (A) no change in tumour cellularity; (B) a strong reduction in tumour cellularity post-treatment only; (C) dramatic reductions in tumour cellularity mid- and post-treatment; and (D) a temporary reduction in tumour cellularity mid-treatment only. Patients for whom the data was incomplete were placed in Group E. When all patients were assessed simultaneously, statistically significant reductions in tumour cellularity were observed mid- and post-chemotherapy.
Given the strong correspondence between the patterns of change in RIN values and tumour cellularity values during treatment, the tumour RIN values were assessed to see if the changes in RIN were an accurate reflection of treatment response in patients (as measured by changes in tumour extent or cellularity). The relationship between the percentage of tumour cells in core biopsies and the maximum RIN value for core biopsies pre-, mid-, and post-treatment was analyzed and summarized in Table 5.
0.0003
0.004
0.001
0.004
As shown in Table 5, there was a very strong positive relationship between tumour extent values and RIN values post-treatment (p values ranged from 0.0003 and 0.004, depending upon whether Spearman or Pearson coefficients were computed and whether the data was transformed to improve symmetry and stabilize variances). The strong correlation between RIN values and patient response to chemotherapy (as measured by tumour cellularity levels) was not observed mid-treatment, possibly because the effects of the chemotherapy agents on tumour RIN had not been fully realized at the mid-point of treatment.
Photomicrographs of sections of tumours pre-, mid-, and post-treatment from patients from each of Groups A (CAMN-006; CAMN-018), B (CAMN-009), C (CAMN-047) and D (CAMN-030) are shown in
The collected RIN values were examined to see whether post-treatment maximum tumour RIN values of ≦3.1 could accurately identify patients whose tumours were responding to chemotherapy (as measured by post-treatment tumour cellularities to ≦10%). A RIN value of 3.1 was selected, as it represented half of the mean RIN value of all pre-treatment tumour core biopsies except those in Group E (see Table 4, above). As shown in Table 6 (see below), patients that had a mean post-treatment tumour RIN value of 3.1 or less also had a post-treatment tumour cellularity of <10% in 19 of 20 patients (95% agreement). Moreover, 16 of 25 patients that had post-treatment tumour RIN values of >3.0 has post-treatment tumour cellularity values of ≧10% (64% agreement). Thus, given specific cut-off values, there was a good correlation between post-treatment RIN values and post-treatment tumour cellularity values, in particular for responders. However, discrepancies in this relationship occurred in four instances where tumour cellularity levels were observed to be high (≧50%) and RIN values were zero. Because breast tumours are known to be heterogeneous, some regions of the tumour may have high tumour cellularity, while other regions do not, thus resulting in discordance between tumour RIN and tumour cellularity. Alternatively, this may be due to tumour cells retaining their cellular morphology but being nonviable with completely hydrolyzed RNA. Therefore, tumour RIN measurements may be superior to tumour cellularity measurements in determining tumour response to chemotherapy agents, since tumour RIN serves as a functional biomarker.
aReduction in maximum RIN post-treatment to ≦3.1 correctly identified responders in 19 of 20 cases (concordance was 95%);
bMaximum tumour RIN values above 3.1 correctly identified non-responders in 16 of 25 cases (64% accuracy);
cDiscrepancy between maximum RIN and tumour cellularity.
The relationship between tumour RNA integrity and actual clinical response to disease was examined, as it was considered an important measure of the utility of tumour RNA integrity to serve as a biomarker of response to chemotherapy agents.
Patient response to chemotherapy treatment was measured in a variety of ways. Patients were deemed to have a complete clinical response (CR) if no tumours were evident following treatment. Tumours whose volume decreased by greater than 50% were deemed to have a partial response (PR) to therapy. Patients with tumours that exhibited no change in size were classified as having stable disease (SD), while patients with new tumours or whose tumours increased in size were said to have progressive disease (PD). Patients who had a complete resolution of disease microscopically were deemed to have exhibited a complete pathologic response (pCR). Using the above definitions of patient response to chemotherapy, the relationship between average or maximum tumour RNA values and clinical response to therapy was analyzed. Both average and maximum RIN values were computed, since one or both may have the greatest correlated with clinical response. In the MA.22 patients, 13 patients were observed to have CRs, while 37 patients were deemed nonresponders (PR, SD, or PD). Only 7 of the 50 patients exhibited a pCR. In particular, a low average tumour RIN post-treatment was significantly associated with a CR (p=0.01), while a low maximum RIN was associated with a pCR mid-treatment (p=0.01), but not post-treatment (p=0.28).
It was noted that while a strong reduction in the maximum tumour RIN was observed mid-treatment for patients exhibiting pCRs post-therapy, tumour RIN values increased or stayed the same, despite complete resolution of disease microscopically. After initial decreases in RIN were observed mid-treatment, tumour RIN values increased post-treatment for a number of MA.22 patients. This may be due to the fact that since breast tumours are heterogeneous, the measured tumour RIN reflected the RNA quality of all cells comprising the tumour, including any normal breast tissue and other cell types. As shown in Table 4, the tumour cellularity of the vast majority of patient tumours pre-treatment was very high (90.94±2.17).
As provided above, strongly reduced RIN values observed mid-treatment reflected a loss of RNA and RNA quality specifically in tumour cells. In view of the results, it would be expected that a low tumour RIN value would correlate with pCR mid-treatment. However, upon clearance of the chemotherapy drugs from the circulatory systems of patients post-treatment, it is possible that normal breast and other tissues may infiltrate the lesion in responders to therapy that no longer have detectable tumour cellularity. Hence, RNA isolated from lesions post-treatment may stem from both normal cells and tumour cells and a high tumour RIN value post-treatment may not necessarily indicate a recurrence of disease. This may explain why the relationship between low maximum tumour RIN values and pCR is most significant when patients are assessed mid-treatment. Furthermore, this may be why tumour RIN values post-treatment are more reliable at predicting responders to epirubicin/docetaxel chemotherapy than nonresponders. Should tumour biopsies post-treatment include non-tumour cells, which would be expected, then patients responding to epirubicin/docetaxel chemotherapy (in particular, a pCR) would be those patients that exhibit a strong reduction in tumour RIN mid- OR post-treatment.
According to the method of Example 1, RNA is extracted from tumour cells of a cancer patient with one or more tumours at two or more different time points during the administration of a chemotherapy regimen, before the administration of a chemotherapy regimen, and during and/or after completion of the regimen.
The tumour cells are collected in one or more image-guided biopsies. An image-guided biopsy is obtained with image-guided means such as computed tomography (CT), x-ray, ultrasound, and magnetic resonance imaging (MRI).
The quality of the extracted RNA is then determined by capillary electrophoresis of the extracted RNA and quantification of the RNAs in the resultant electropherogram. An automated analytical system, such as the Agilent® 2100 Bioanalyzer (Agilent Technologies, Inc., U.S.A.) is used for carrying out the RNA quality determination, in order to obtain an RNA integrity number (RIN) for each sample of RNA (Schroeder, A. et al., BMC. Mol. Biol. 7 (2006) 3; Imbeaud, S. et al. Nuci. Acids Res. (2005), 33, 6, e56, 1-12).
The RIN value of the tumour cells collected before administration of the chemotherapy regimen is then compared with the one or more RIN values of tumour cells collected after commencement of the regimen, i.e. during the regimen and/or after completion of the chemotherapy regimen. If a patient exhibits no change in tumour RNA integrity during treatment (response pattern A as noted in Example 2), then the patient's tumour would be considered resistant to the chemotherapy regimen being used. The patient would be considered at high risk of tumour progression and prognosis would be considered poor. Alternative chemotherapy regimens or treatment protocols should then be considered. The method outlined herein can be repeated to determine responsiveness to the new regimen.
If a patient exhibits a dramatic reduction in tumour RNA integrity (>50%) both mid- and post-treatment (response pattern C as noted in Example 2), then the patient would be considered to have responded to chemotherapy and would be at lower risk of tumour progression. The patient's prognosis would be considered good. Tumour RIN values near zero would be highly indicative of response to chemotherapy and low risk of tumour progression.
If a patient exhibits a dramatic reduction in tumour RNA integrity post-treatment only (response pattern B as noted in Example 2), then the patient would be considered to have responded to chemotherapy and be at lower risk of tumour progression. The patient's prognosis would be considered good.
If a patient exhibits a dramatic change in tumour RNA integrity mid-treatment only, then she likely has responded to therapy and would be at a lower risk of disease recurrence. This is regardless of a return to high “tumour” RNA integrity post-treatment, since the high quality RNA post-treatment may stem from normal tissue that has infiltrated the lesion. However, in this case, it is possible that the tumour has recurred post-treatment.
Numerous modifications, variations, and adaptations may be made to the particular embodiments of the invention described above without departing from the scope of the invention which is defined in the following claims.
This application is a continuation of U.S. application Ser. No. 12/676,815 filed on Sep. 3, 2010, which is a National stage entry of International Application No. PCT/CA2008/001561 filed on Sep. 5, 2008, which claims priority to U.S. Provisional Applications No. 60/935,903 filed Sep. 6, 2007 and U.S. Provisional Application No. 60/935,874 filed Sep. 5, 2007, each of these applications being incorporated herein by reference in their entirety.
| Number | Date | Country | |
|---|---|---|---|
| 60935903 | Sep 2007 | US | |
| 60935874 | Sep 2007 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 12676815 | Sep 2010 | US |
| Child | 14951049 | US |
