SEARCH

SEARCH BY CITATION

Keywords:

  • biomarker;
  • canine;
  • chemotherapy;
  • lymphoma;
  • thymidine kinase

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Assay of thymidine kinase activity
  6. Statistical analysis
  7. Results
  8. Discussion
  9. References

The aim of the study was to evaluate if thymidine kinase (TK) correlated with duration of first remission (DFR) or survival in dogs with lymphoma and if initial TK levels correlated with stage and substage; and also to assess if TK level at diagnosis correlated with immunophenotype. TK was assayed in 73 dogs with treatment naïve lymphoma, then again after treatment; 47% had an initial TK above the reference interval. Dogs with B-cell lymphoma had higher initial TK activities than dogs with T-cell lymphoma. TK levels were not higher in dogs with higher stage disease and TK activity prior to treatment was not associated with DFR or survival. Where TK was elevated at diagnosis, it fell into the reference range during remission. TK was normal in 53% dogs at diagnosis, which is higher than previously reported. Further studies are warranted to assess the utility of TK in dogs with lymphoma.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Assay of thymidine kinase activity
  6. Statistical analysis
  7. Results
  8. Discussion
  9. References

Lymphoma is responsible for approximately 5% of all malignant canine cancers with an estimated annual incidence of approximately 25 per 100 000 dogs.1 Although diagnosis is rarely problematic, predicting prognosis, objectively monitoring remission status and detecting relapse promptly in treated canine lymphoma patients remains challenging.

Established prognostic indicators for canine lymphoma include clinical stage, WHO substage, histologic grade, immunophenotype, prior corticosteroid exposure and certain anatomic locations.2 A host of other factors have been investigated as potentially useful prognostic indicators. These include various markers of proliferation, including argyrophilic nucleolar organizing regions (AgNOR staining)3 and Ki-67 antibody staining.4 Serum levels of circulating endothelial markers and matrix metalloproteinases may also be prognostically significant, and serum lactate dehydrogenase (LDH) levels also have been investigated as prognostic indicators in lymphoma and other malignancies, but do not appear to be useful for the majority of individual patients.5

Evaluation of remission status is also prognostically important. Minimal residual disease (MRD), defined as residual malignant cells after antineoplastic treatment, is a potential source of relapse in many cancer states.6 Recently, MRD was detected in human lymphoma patients in clinical remission, using a real-time polymerase chain reaction (PCR)-based quantification system.7 In human haematological malignancies, the quantity of MRD after treatment has been proposed as an indicator for treatment outcome.7

In addition to PCR-based detection of MRD, biomarkers are utilized in human diseases to monitor remission status.5,8 In dogs with lymphoma, serum thymidine kinase (TK)9,10 and serum alpha 1 acid glycoprotein11 concentrations have been shown to predict relapse before it is clinically confirmed in a some cases. Gaining useful prognostic and monitoring information by obtaining a simple blood sample, as with TK, is extremely attractive. Biomarkers such as TK would also be ideal to use for monitoring remission status in dogs without peripheral nodal involvement, as their disease cannot be readily assessed on clinical examination and repeated imaging to assess remission status and response to treatment is expensive.

The TK is a cytoplasmic enzyme that catalyses the phosphorylation of thymidine to thymidine monophosphate. It exists in two forms: cytoplasmic thymidine kinase 1 (TK-1) and mitochondrial thymidine kinase 2 (TK-2). TK-1 is associated with cellular proliferation, whereas TK-2 is needed for mitochondrial DNA precursor synthesis. TK-1 activity increases markedly after the G1-S transition in the cell cycle and then declines rapidly in G2. Any increase in TK-1 activity extracellularly may thus reflect an overall increase in DNA synthesis and the number of cells dying in the replicative stage of the cell cycle and releasing TK-1 to the blood.8,12

Many haematopoietic malignancies are characterized by a very high rate of cell proliferation, reflected by altered serum thymidine kinase activity (TK, actually TK-1). TK-1 has been found to provide information regarding prognosis and treatment response in humans with leukaemia, multiple myeloma, Hodgkin's lymphoma, myelodysplasia and non-Hodgkins lymphoma.8,13 Furthermore, TK has been used to distinguish progressive and indolent disease in chronic lymphocytic leukaemia in people.14–18 TK analysis also allows a more precise risk classification in human large cell lymphoma.8 Recently, serum thymidine kinase activity has been suggested as a potent prognostic indicator and disease monitoring tool in canine lymphoma.9,10

Lymph node palpation is a crude method of assessing remission status in dogs with lymphoma. The return of initially high serum TK activity to within the reference interval after chemotherapy may give more objective evidence of successful therapy than clinical remission status based on lymph node palpation alone. This would give the attending clinician more confidence to withdraw therapy. However, if the values were still elevated, this might indicate a likelihood of rapid relapse, and continued therapy or a change in therapy may be warranted.

The initial objectives of this study were to test previously published observations by investigating if:

  • 1
    TK could be used as a prognostic marker for survival time in dogs with multicentric lymphoma and if initial TK levels correlated with tumour stage and immunophenotype.
  • 2
    TK activity normalized on induction of remission with chemotherapy, in dogs where TK activity was elevated prior to therapy.
  • 3
    Normal TK activity at cessation of non-continuous chemotherapy was associated with overall longer remission time.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Assay of thymidine kinase activity
  6. Statistical analysis
  7. Results
  8. Discussion
  9. References

Between January 2007 and March 2010, serum samples were collected from 73 dogs presented at the Small Animal Teaching Hospital, University of Liverpool with high-grade treatment naïve lymphoma, confirmed either cytologically or histologically. One further dog (dog 74 in Fig. 1) did not have TK measured prior to therapy, but when confirmed to be out of remission after receiving induction treatment. The grade of the lymphoma was reported as ‘high’ by the examining pathologist, usually following a description of large neoplastic cell size and brisk mitotic activity. Immunohistochemistry was performed in some cases (where formalin-fixed tissue was present), or immunocytochemistry of lymph node aspirates. Information obtained and recorded from patients' medical records included gender, breed, age and neuter status, results of clinical staging, treatments administered, response to treatment, duration of first remission (DFR) and survival time. Animals that had received prior treatment with systemic corticosteroids or any chemotherapy agent were excluded.

imageimage

Figure 1. Table showing demographics for 74 dogs with multicentric lymphoma. Immunophenotype is shown where data is available. The values of TK prior to treatment (TK-pre) and following instigation of treatment (TK-post) are shown. Survival time for each dog is recorded and whether the animal was alive at the time of writing or dead. A ‘-’ denotes that data was not available because staging or immunophenotyping was declined by the owners. TK-pre was not measured in dog 74 pre-treatment: only when disease had not clinically resolved after initial therapy. Bernese = Bernese Mountain Dog, CKCS = Cavalier King Charles Spaniel, Golden = Golden Retriever, JRT = Jack Russell Terrier, Lab = Labrador Retriever, OES = Old English Sheepdog, SBT = Staffordshire Bull Terrier, Springer = English Springer Spaniel, Std poodle = Standard poodle, Weim = Weimeraner, WHWT = West highland white terrier, XB = Cross breed, Yorkie = Yorkshire Terrier.

A complete blood cell count and serum biochemistry was performed in all cases at presentation, as well as urine analysis in most cases. In all cases, smear examination was carried out to provide a manual differential count, reticulocyte count and to evaluate cellular morphology. Radiography of the thorax and abdominal ultrasonography were performed for clinical staging purposes. Additional staging procedures, such as bone marrow evaluation, were performed in some cases as deemed appropriate by the attending clinician. Staging was based on the modified World Health Organization (WHO) staging system for canine lymphoma (Table 1). Animals were also classified as WHO substage ‘a’ or ‘b’ depending on whether they were clinically well or suffering from clinical signs of their lymphoma at presentation, respectively.

Table 1.  Modified WHO staging criteria for canine lymphoma
Stage 1Disease confined to a single lymph node
Stage 2Regional lymphadenopathy (confined to one side of diaphragm)
Stage 3Generalized lymphadenopathy
Stage 4Hepatosplenomegaly (with or without lymphadenopathy)
Stage 5Bone marrow, CNS, or other extranodal site involvement
Substage a: No clinical signs 
Substage b: Clinical signs of illness 

Complete remission (CR) was defined as complete regression of measurable tumour. Partial response (PR) was defined as >50% but <100% regression of measurable tumour. Progressive disease was defined as an increase of >25% of measurable tumour. Stable disease (SD) was defined as an increase in size of <25% or >50% regression of measurable tumour. This data was recorded in the clinical notes. Any change in nodal consistency, e.g. increased firmness of any peripheral nodes, was also recorded.

After completion of discontinuous chemotherapy (where applicable), physical examinations were performed monthly until relapse. If this occurred, it was advised that another cycle be given. Repeated discontinuous cycles were given until lack of tumour response or owner wishes. Other commonly used rescue protocols included DMAC19 (dexamethasone, melphalan, actinomycin D and cytarabine) and lomustine. Some dogs received methotrexate, chlorambucil, further doses of l-asparaginase, mitoxantrone, doxorubicin and temozolomide.20

Serum was taken for TK measurement in 73 dogs prior to any treatment. Blood was placed in serum gel tubes and posted so that it arrived at the laboratory within 24 h. On arrival it was frozen (−20 °C) and samples were run in batches every 7–14 days depending on the number of samples. One dog did not have TK measured prior to treatment, but after the first few weeks of chemotherapy when its lymph nodes were still moderately enlarged. If TK activity was elevated (>7 U L−1) prior to chemotherapy, then it was measured again at either: (1) the end of a 25-week non-continuous CHOP-based protocol; (2) when considering withdrawing COP-based chemotherapy or (3) when remission status was uncertain based on clinical examination during any treatment protocol or any period of monitoring after chemotherapy had been discontinued. If serum TK activity was within-range (<7 U L−1) prior to initiation of treatment, then it was not rechecked at a later time, as there is currently no known value in the monitoring of TK activity in this group of patients, and this was a non-funded clinical trial. For survival analysis, animals were censored if they were still alive at the time of data accrual.

Assay of thymidine kinase activity

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Assay of thymidine kinase activity
  6. Statistical analysis
  7. Results
  8. Discussion
  9. References

The TK activity in serum was determined using a 125I-based radioenzyme assay as previously described.9 The reference interval (normal 0–7 U L−1) was derived from a previous veterinary study.9

Statistical analysis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Assay of thymidine kinase activity
  6. Statistical analysis
  7. Results
  8. Discussion
  9. References

Data were entered into a spreadsheet (Microsoft Excel 2007, Microsoft Ltd, Microsoft UK, Reading, UK) and checked and corrected as required. Statistics were performed using the package Minitab® version 15.1.1 (MINITAB Ltd, Coventry, UK). After performing basic descriptive statistics, univariable comparisons employed the nonparametric Mann–Whitney and Wilcoxon Signed-Rank tests. TK values were logged for graphical presentation. Basic survival distributions were examined using the Kaplan–Meier method and comparisons tested with the Log-Rank test. The critical probability was defined as P < 0.05. For the purpose of the statistical analysis, when testing for differences between TK activity pre- and post-treatment, if a dog had more than one post-treatment TK measured while in CR, then the median value was used. Dogs still alive at the time of survival analysis were censored. Multivariate analysis was performed with the use of forward stepwise multiple logistic regression to the Cox proportional hazards model for survival.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Assay of thymidine kinase activity
  6. Statistical analysis
  7. Results
  8. Discussion
  9. References

Seventy nine dogs had TK activity measured prior to therapy. Six dogs were excluded as their disease was not multicentric. These included two dogs with epitheliotropic lymphoma, two dogs with diffuse gastrointestinal lymphoma, one dog with pericardial and skin lymphoma and one dog with vaginal and skin lymphoma.

The most commonly encountered breeds were cross-breed dogs (16/73, 22%), Boxers (6/73, 8%), Golden Retrievers (5/73, 7%) and Dogue de Bordeaux and Labrador retrievers (both 4/73, 5%). The mean and median ages were 6.4 and 6.5 years, respectively (range 1–14 years). There were 41 male and 32 female dogs. There was 1 dog with stage 1 disease, 2 dogs with stage 2 disease, 13 dogs with stage 3 disease, 34 dogs with stage 4 disease, 16 dogs with stage 5 disease and 7 dogs in which staging was not undertaken because of financial reasons or the owners' wishes. Thirty two dogs were classified as WHO substage ‘b’ and the remainder substage ‘a’. Most of the dogs were treated with chemotherapy protocols. Sixty three animals received a modified Madison–Wisconsin 25-week discontinuous21 (CHOP-based) protocol (Table 2) initially, and six animals received a COP-based protocol (Table 3). One dog received one dose of l-asparaginase only, one dog prednisolone only and two dogs were not treated as this was declined by the owner(s).

Table 2.  The modified Madison–Wisconsin chemotherapy protocol used in most of the treated dogs with multicentric lymphoma
Week12345678910
Vincristine (0.7 mg/m2 i.v. )* *  * *  
Prednisolonea (mg/kg)21.510.5      
l-asparaginase (400 iu/kg i.m.)*         
Cyclophosphamide (250 mg/m2 i.v.) *    *   
Epirubicin (30 mg/mg2 i.v.)   *    * 
Week1113151719212325
  1. Epirubicin is used as the standard anthracycline at the authors' institution.

  2. aPrednisolone 2 mg/kg sid p.o. week 1, then 1.5 mg/kg sid week 2, then 1 mg/kg week 3, then 0.5 mg/kg week 4.

Vincristine (0.7 mg/m2 i.v. )* * * * 
Cyclophosphamide (250 mg/m2 i.v.) *   *  
Epirubicin (30 mg/mg2 i.v.)   *   *
Table 3.  A high-dose COP protocol used in six of the dogs treated with multicentric lymphoma
Week1234567
  1. Week 7 is then repeated continually every third week until remission is lost or chemotherapy is withdrawn.

Vincristine (0.7 mg/m2 i.v. )****  *
Cyclophosphamide (250 mg/m2)*  *  *
Prednisolone 1 mg/kg PO every other day*******

Seventy three dogs had TK activity measured initially prior to any treatment. Initial TK activity demonstrated a high degree of variation (median 6.2; range <1–148; normal reference range 0–7 U L−1). Thirty four dogs (47%) had an initial TK above the reference range (median 25.5; range 7.1–148 U L−1; Fig. 1). In the one extra dog that had TK measured after an incomplete response to treatment, it was markedly elevated at 51 U L−1.

Forty six (63%) dogs had immunophenotyping performed. Thirty two dogs (70%) had B-cell lymphoma and 14 dogs (30%) had T-cell lymphoma. Seventeen of 32 dogs (53%) with B-cell lymphoma had elevated TK pre-treatment; 4 of 14 dogs (29%) with T-cell lymphoma had elevated TK pre-treatment. Dogs with B-cell lymphoma had higher initial TK activities than dogs with T-cell lymphoma (medians 8.1 and 3.7 U L−1, respectively) and this difference was statistically significant (inline image; Mann–Whitney test; Fig. 2).

image

Figure 2. Log of pre-treatment TK (Pre-TK) values for dogs with T-cell (inline image) and B-cell (inline image) lymphoma, showing a statistically significantly higher pre-treatment TK values for dogs with B-cell lymphoma compared to dogs with T-cell lymphoma.

Download figure to PowerPoint

There was no association between initial TK activity and clinical stage (inline image; Wilcoxon Signed-Rank test). Only dogs in stages 3–5 were included in the statistical analysis because of the low numbers of dogs present with stage 1 and 2 disease. Other than one dog with stage 3 disease (dog 73 in Fig. 1), no dogs with stages 1–3 disease had an extremely high TK value.

Eleven dogs that had elevated TK activity pre-treatment had TK remeasured while in CR based on clinical examination. Some dogs had more than one sample taken, and samples were taken at various time points (i.e. during 25-week discontinuous therapy or during continuous chemotherapy, or after completing discontinuous treatment), resulting in 18 samples from these 11 dogs (Table 4). All of these ‘post-treatment’ (i.e. samples taken after the initiation of chemotherapy, while the animal was considered to be in remission clinically). TK values were within the normal reference range. TK activity prior to treatment was significantly higher than TK activity after treatment (inline image; Mann–Whitney test; Fig. 3). Ultimately, only 7 of these 11 dogs had TK activity measured again at the end of a discontinuous protocol. In all of these dogs TK activity was elevated at diagnosis and was within the reference interval at the end of the protocol, when the dogs were in remission based on clinical examination. Therefore, it could not be assessed whether dogs in this situation (i.e. when TK activity has normalized at the end of a finite chemotherapy protocol) have improved survival times.

Table 4.  Eleven dogs with elevated TK prior to treatment had TK measured again post-treatment (four dogs had multiple measurements post-treatment) while in complete clinical remission (CR)
DogTK prior to treatment (U L−1)TK post-treatment #1 (U L−1)TK post-treatment #2 (U L−1)TK post-treatment #3 (U L−1)TK post-treatment #4 (U L−1)TK post-treatment #5 (U L−1)
  1. All samples taken post-treatment were within the reference interval.

1 12.71    
2 7.12.3    
3 7.73.43.4   
4 13.83.51.61.211
51191.2    
61215.31.7   
7 962.2    
8 201.21.7   
9 331    
10 483.9    
11 7.81.2    
image

Figure 3. Line plot showing elevated TK values in 11 dogs falling to within-range following instigation of chemotherapy.

Download figure to PowerPoint

Four dogs that had elevated TK activity pre-treatment, and one dog that did not have TK measured at initial diagnosis, had TK measured when a lymph node(s) was becoming larger or firmer and loss of remission was suspected on clinical examination. All five of these dogs had elevated TK values (>7 U L−1) when considered not in remission (Table 5). Dogs 1 and 2 had a TK that was persistently elevated after treatment had been initiated and not in CR, but this returned to reference interval in both cases when they were rescued with alternative chemotherapy agents and subsequently deemed to be in CR. For dog 3, third post-treatment TK measurement was elevated which corresponded to peripheral lymph node enlargement and relapse. This dog had been shown to have a normal TK on two prior occasions when in clinical remission. Dog 4 had an elevated TK prior to treatment and initially responded well to chemotherapy. However, shortly into protocol the dog developed lymphadenopathy again and TK was shown to be elevated at this time point.

Table 5.  Table showing elevated post-treatment TK values in five dogs, when it was measured when they were considered out of remission clinically
DogTK prior to treatment (U L−1)TK post-treatment #1 (U L−1)TK post-treatment #2 (U L−1)TK post-treatment #3 (U L−1)
  1. Elevated TK values shown in bold.

17.8 9.5 1.71.2
2121 16.6 5.31.7
37.73.43.4 17.4
420 45   
5Not measured 51   

Overall median DFR and survival times for all dogs with lymphoma were 238 and 324 days, respectively. TK activity prior to treatment was not associated with DFR (inline image, Wilcoxon Signed-Rank test) or survival time (inline image; Wilcoxon Signed-Rank test). The dogs were split into quartiles (1, 2, 3 and 4) depending on their increasing level of TK pre-treatment. Figure 4 shows that there was no significant difference in survival between the groups. Sixteen dogs had a TK >30 U L−1 prior to treatment. Dogs with a TK-pre value >30 U L−1 did not have a significantly poorer DFR (inline image; Wilcoxon Signed-Rank test) or survival (inline image; Wilcoxon Signed-Rank test, Fig. 5) compared to the dogs with a pre-TK <30 U L−1.

image

Figure 4. Kaplan–Meier survival plot. Dogs were placed in one of four quartiles depending on the level of TK prior to treatment, with dogs in quartile 1 having the lowest TK values prior to therapy and dogs in quartile 4 having the highest TK values prior to therapy. There was no significant difference between the quartiles in terms of survival. Quartile 1 = TK 1.0 − 2.5  U L−1 (19 dogs), quartile 2 = TK 2.8 − 6.2 U L−1 (18 dogs), quartile 3 = TK 6.8 − 20 U L−1 (19 dogs) and quartile 4 = 31 − 148 U L−1) (17 dogs).

Download figure to PowerPoint

image

Figure 5. Kaplan–Meier plot showing no difference in the survival between dogs with pre-treatment TK values <30 U L−1 compared with dogs with a pre-treatment TK value >30 U L−1.

Download figure to PowerPoint

DFR and survival times were significantly decreased for dogs with T-cell lymphoma compared to dogs with B-cell lymphoma (Fig. 6) and for animals hypercalcaemic at presentation compared to those not hypercalcaemic at presentation. Dogs classified as WHO substage ‘a’ had superior DFR and overall survival times compared to substage ‘b’ (Table 6). There was no statistical association between WHO substage and level of TK at diagnosis (inline image; Kruskal–Wallis test). There was no association between age (inline image; Log-Rank test), breed (inline image; Log-Rank test) or sex (inline image; Log-Rank test) and survival. On multivariate analysis, only immunophenotype retained prognostic value for duration of first remission (inline image) and overall survival (inline image).

image

Figure 6. Kaplain–Meier survival curve showing poorer survivals for the dogs with T-cell lymphoma compared to dogs with B-cell lymphoma.

Download figure to PowerPoint

Table 6.  Table revealing the variables shown to be prognostic after univariable analysis in the dogs treated for lymphoma
Prognostic factorMedian DFR (days) P valueMedian survival (days) P value
  1. Hypercalcaemia, WHO substage ‘b’ and T-cell lymphoma are shown to carry a poorer prognosis. All comparisons were tested with the Log-Rank test. DFR = duration of first remission.

WHO substage     
a3670.009446<0.001
b135 163 
Hypercalcaemia at diagnosis     
Y1360.029186<0.005
N280 324 
Immunophenotype     
B3520.029408<0.001
T125 136 

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Assay of thymidine kinase activity
  6. Statistical analysis
  7. Results
  8. Discussion
  9. References

Studies have suggested that the majority, or all, dogs with lymphoma have TK activity above the reference range.9,22 In this study, only 45% of dogs had an elevated TK at presentation. This is a significant finding, as more than half of the study population had TK activity within reference interval, and using TK activity to prognosticate about survival and to monitor these patients' clinical remission status is unlikely to be useful. In human medicine there are several subtypes of non-Hodgkins lymphoma (NHL)23 and this has also been shown in veterinary medicine.24–27 At least one veterinary study has shown significant prognostic differences between the clinic-morphological subtypes of canine NHL25 and these observations are suggestive of differing inherent biological features and clinical behaviour of these lymphoma entities. It may be that TK activity is most useful when applied within a particular histologic and immunophenotypic subclassification of lymphoma. It is possible there may be a significant difference between the canine lymphoma population in the previous studies and the United Kingdom population. Another possible explanation for the smaller number of dogs with an elevated pre-treatment TK is that the dogs with lymphoma in this study were diagnosed at an earlier time point in the course of their disease, with a resulting lower tumour burden and thus smaller overall global DNA synthesis. However, this would not explain some dogs that presented with significantly bulky disease and pre-treatment TK well within the reference interval. It would be interesting to examine TK values from dogs with so-called indolent lymphomas. These tumours are not diagnosed as frequently in veterinary medicine and are characterized as having a low mitotic rate and slow clinical progression with a good long-term prognosis.28 For this reason, these dogs may be expected to have a lower TK at diagnosis compared to dogs with the more commonly diagnosed high-grade lymphomas. However, it seems unlikely that a significant number of the dogs in this study had indolent lymphomas, as the pathologist reported a cytologically or histologically high-grade tumour and also many of the dogs with low pre-treatment TKs had poor survivals (Fig. 1) which would not be characteristic of an indolent disease.

The canine reference range (<7 U L−1) is based on published data, and was determined by measuring TK activity in 21 healthy adult dogs by the same methods used in this study.9 Further studies may be indicated to validate if the upper end of the reference range is an appropriate cut-off value for monitoring remission in treated patients, particularly considering the large number of dogs in this study with initial TK values below this limit. It would be useful to measure post-treatment (in remission) serum TK levels in a cohort of dogs with a pre-treatment TK that was not elevated (i.e. not >7 U L−1) to see if the value decreases as this may still be potentially useful (i.e. if TK activity declines with successful treatment of the disease). New cut-off values would need to be defined for this to be clinically useful. It would have been helpful to measure TK in a large group of healthy control animals in the hospital, however, financial limitations made this impossible in this study. Practical use of tumour biomarkers is often hampered by their lack of specificity, generally because of measurable levels of the marker in normal individuals or those with benign disease.29–32 Therefore, the upper end of the reference interval must be sufficiently high to avoid an unacceptable false positive rate.33 A false positive result, or wrongly diagnosing an owners pet with neoplastic disease, may have distressing consequences. In the context of TK, false positives may result in unnecessary instigation of a change in chemotherapy regime or initiation of rescue chemotherapy with associated potential negative effects on quality of life and financial implications. Alternatively, an upper limit that is too high may not prompt a change in chemotherapy or restarting chemotherapy until the disease has progressed further which may adversely affect outcome. A better reference interval for canine TK will likely result following sampling of a much larger number of healthy control dogs and studying of a greater number of patients that have had serial sampling performed.

The TK enzyme has previously been shown to be stable for 24 h at either 37 °C or room temperature and for up to 3 months at −20 °C.9 Therefore based on the methods of sample handling in this study, results would not be expected to have been significantly affected by freezing and running samples in batches.

The median and mean ages of the dogs in this study were 6.5 and 6.6 years, respectively (1–14 years), and this is similar to previous work.21 While serum TK activity has been shown to be significantly higher in young dogs compared to adults,9,10 due to the higher cellular proliferation rate occurring in immature dogs,18 there is no work determining if younger mature dogs have higher TK values than older animals. However, the three youngest dogs in this study were 1-year old (one dog) and 2-year old (two dogs) and all had TK activity within the reference interval. There was no association between age and survival in this study.

In this study, dogs with B-cell lymphoma had higher pre-treatment TK activity than dogs with T-cell lymphoma. Only 55% of dogs with B-cell lymphoma and 30% of dogs with T-cell lymphoma had elevated initial TK activity prior to therapy. Dogs with B-cell lymphoma also had significantly higher pre-treatment TK values than dogs with T-cell lymphoma. As far as it is known this finding has not been reported in the veterinary literature. The reason for this observation is unclear and may represent a difference in the reliance on this enzymatic pathway in neoplastic B and T cells. There was no significant difference in the number of dogs in each WHO stage group between dogs with B- and T-cell lymphoma which suggests the difference is unlikely to be because of a difference in tumour burden.

Of the 18 samples from 11 dogs that had TK activity measured during or after treatment, while in CR on clinical examination, all values were within the normal reference range. This is logical as TK activity correlates with the global proliferative activity of tumour cells and will decline with successful therapy. The data support previous work suggesting that where dogs have an elevated TK activity before commencement of treatment, it declines to physiological activity once clinical remission has been achieved.9 One drawback in this study is that most dogs were not restaged with repeat lymph node aspirates, thoracic radiographs and abdominal ultrasound at the time of repeat TK analysis to confirm CR status. Peripheral lymph node palpation is a crude method of documenting response to treatment and remission status, as thoracic/abdominal nodes and parenchymatous organs may still be grossly affected which may not be detected on physical examination. The five dogs which had lost remission clinically had elevated TK at this time, supporting previous observations that TK levels become elevated again when remission is lost.18 However, this is a very small number of dogs and more dogs would need to be followed to corroborate this observation.

Von Euler et al. reported that TK activity correlated significantly to the clinical stage of disease.9 This was not seen in this study. Of particular interest were the four dogs [dogs 10 (stage 5), 23 (stage 4), 24 (stage 4) and 28 (stage 4) in Fig. 1] with subjectively significantly bulky, high-grade disease that had TK activities within the reference range before treatment (pre-treatment TK of 1.6, 3.5, 3.8 and 4.4 U L−1, respectively). In common with other studies, it is very difficult to accurately differentiate between dogs with stage 3 and 4 lymphoma as fine needle aspirates were not routinely performed. This is because the treatment is the same for dogs with either stage 3 or 4 disease. Of note, is that fact that other than one dog with stage 3 disease (dog 73 in Fig. 1) no dogs with stage 1–3 disease had an extremely high TK value. Perhaps a larger number of dogs may disclose a more subtle difference in pre-treatment TK levels and clinical stage.

The dogs in this study that were hypercalcaemic at presentation had a significantly poorer DFRs and survivals than the dogs that were not hypercalcaemic at presentation. There is a well-accepted association between hypercalcaemia and survival in canine lymphoma patients. However, this is invariably not significant with multivariate analysis because of the association with T-cell immunophenotype, which in itself is strongly associated with a negative prognosis, as also found in this study.2,34

Previous data has shown that for patients with high (>30 U L−1) versus low (<30 U L−1) pre-treatment TK, the 50% survival time was 1 month and 9 months, respectively.9 There was no such correlation with survival time or indeed DFR in this study indicating that initial TK activity >30 U L−1 may not be such a potent prognostic indicator in dogs with lymphoma and needs to be further evaluated. Indeed, three dogs among those with the highest TK values prior to treatment [dogs 66 (inline image), 71 (inline image) and 72 (inline image)] had prolonged survivals of 628, 782 and 587 days, respectively (Fig. 4). While contradicting previous data, perhaps this observation reflects the fact that response to therapy is a very important factor in determining overall patient prognosis. Future studies should consider analyzing the rapidity by which TK levels reduce to within-range and assessing for any correlation with prognosis (i.e. perhaps the prognosis is more favourable in patients where TK falls more rapidly to within the reference interval). If correct, perhaps this could be considered to be the best use of a tumour biomarker; namely to follow the patients' own response using their own initial pre-treatment value as an internal reference. Further studies would be required to assess this, particularly in a group of dogs with more serial sampling performed.

Duration of first remission times was included in the analyses in this study and this is an important parameter to consider in canine lymphoma. Unfortunately, the nature of lymphoma in the dog is that the majority will relapse and ultimately die of the disease. After initial relapse, there are a multitude of owner choices that can affect the overall survival of the individual patient. Some owners choose to continue with a variety of rescue chemotherapy protocols while others may decide to euthanize immediately at this point for financial or motivational reasons. This parameter is less likely to be affected by these decisions and hence were included in this study.

This study confirms that when dogs with elevated TK activity at diagnosis are treated, enzyme activity levels return reliably to the normal reference range, and this may be an aid in confirming clinical examination findings. However, clearly a large subsection of dogs do not have elevated TK activity prior to treatment and this assay would therefore not be useful for prognostic significance and disease monitoring in these animals if the current reference range is used. This study does not substantiate previous work that suggests initial serum TK activity correlates with survival in dogs with lymphoma.

One major criticism of the study is the limited number of dogs that had TK monitored serially throughout their treatment. Unfortunately, the study was not funded and so repeated TK measurement was at the discretion of the attending clinician.

In humans, early diagnosis and aggressive therapy of non-Hodgkins lymphoma results in a large proportion of patients achieving long-term remission or cure.35 Most dogs with lymphoma have high-grade, advanced-stage disease at diagnosis and even aggressive chemotherapy protocols result in mostly relatively short-term remissions and modest survivals.36 Consequently, particularly when considering the findings of this study, the authors would recommend that the clinical utility, reference range and cost-effectiveness of patient monitoring with TK activity are further studied prior to widespread commercial use, particularly in patients where initial TK is within the reference interval. For patients with elevated TK, identification of suitable time points for monitoring may enhance value.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Assay of thymidine kinase activity
  6. Statistical analysis
  7. Results
  8. Discussion
  9. References
  • 1
    Dorn CR, Taylor DO and Hibbard HH. Epizootiologic characteristics of canine and feline leukemia and lymphoma. American Journal of Veterinary Research 1967; 28: 9931001.
  • 2
    Teske E. Prognostic factors for malignant lymphoma in the dog: an update. Veterinary Quarterly 1994; 16(Suppl. 1): 29S31S
  • 3
    Vail DM, Kravis LD, Kisseberth WC, Ogilvie GK and Volk LM. Application of rapid CD3 immunophenotype analysis and argyrophilic nucleolar organizer region (AgNOR) frequency to fine needle aspirate specimens from dogs with lymphoma. Veterinary Clinical Pathology 1997; 26: 6669.
  • 4
    Renwick MG, Argyle DJ, Long S, Nixon C, Gault EA and Nasir L. Telomerase activity and telomerase reverse transcriptase catalytic subunit expression in canine lymphoma: correlation with Ki67 immunoreactivity. Veterinary and Comparative Oncology 2006; 4: 141150.
  • 5
    Zanatta R, Abate O, D'Angelo A, Miniscalco B and Mannelli A. Diagnostic and prognostic value of serum lactate dehydrogenase (LDH) and LDH isoenzymes in canine lymphoma. Veterinary Research Communications 2003; 27(Suppl. 1): 449452.
  • 6
    Roman J, Alvarez MA and Torres A. Molecular basis for therapeutic decisions in chronic myeloid leukemia patients after allogeneic bone marrow transplantation. Haematologica 2000; 85: 10721082.
  • 7
    Rambaldi A, Lazzari M, Manzoni C, Carlotti E, Arcaini L, Baccarani M, Barbui T, Bernasconi MC, Dastoli G, Fuga G, Gamba E, Gargantini L, Gattei V, Laurea F, Lazzarino F, Mandelli F, Morra E, Pulsoni A, Ribersani M, Rossi-Ferrini PL, Rupolo M, Tura S, Zagonel V, Zaga F, Zinzani P, Reato G and Foa R. Monitoring of minimal residual disease after CHOP and rituximab in previously untreated patients with follicular lymphoma. Blood 2002; 99: 856862.
  • 8
    Hannigan BM, Barnett YA, Armstrong DB, McKelvey-Martin VJ and McKenna PG. Thymidine kinases: the enzymes and their clinical usefulness. Cancer Biotherapy 1993; 8: 189197.
  • 9
    von Euler H, Einarsson R, Olsson U, Lagerstedt AS and Eriksson S. Serum thymidine kinase activity in dogs with malignant lymphoma: a potent marker for prognosis and monitoring the disease. Journal of Veterinary Internal Medicine 2004; 18: 696702.
  • 10
    Von Euler HP, Rivera P, Aronsson AC, Bengtsson C, Hansson LO and Eriksson SK. Monitoring therapy in canine malignant lymphoma and leukemia with serum thymidine kinase 1 activity – evaluation of a new, fully automated non-radiometric assay. International Journal of Oncology 2009; 34: 505510.
  • 11
    Hahn KA, Freeman KP, Barnhill MA and Stephen EL. Serum alpha 1-acid glycoprotein concentrations before and after relapse in dogs with lymphoma treated with doxorubicin. Journal of the American Veterinary Medical Association 1999; 214: 10231025.
  • 12
    Sherley JL and Kelly TJ. Regulation of human thymidine kinase during the cell cycle. Journal of Biological Chemistry 1988; 263: 83508358.
  • 13
    O'Neill KL, Buckwalter MR and Murray BK. Thymidine kinase: diagnostic and prognostic potential. Expert Review of Molecular Diagnostics 2001; 1: 428433.
  • 14
    Ellims PH, Eng Gan T, Medley G and Van Der Weyden MB. Prognostic relevance of thymidine kinase isozymes in adult non-Hodgkin's lymphoma. Blood 1981; 58: 926930.
  • 15
    Ellims PH, Van der Weyden MB and Medley G. Thymidine kinase isoenzymes in human malignant lymphoma. Cancer Research 1981; 41: 691695.
  • 16
    Ellims PH, Gan TE and Van der Weyden MB. Thymidine kinase isoenzymes in chronic lymphocytic leukaemia. British Journal of Haematology 1981; 49: 479481.
  • 17
    Sadamori N, Ichiba M, Mine M, Hakariya S, Hayashibara T, Itoyama T, Nakamura H, Tomonaga H and Hayashi K. Clinical significance of serum thymidine kinase in adult T-cell leukaemia and acute myeloid leukaemia. British Journal of Haematology 1995; 90: 100105.
  • 18
    Musto P, Bodenizza C, Falcone A, D'Arena G, Scalzulli P, Perla G, Modoni S, Parlatore L, Valvano MR and Carotenuto M. Prognostic relevance of serum thymidine kinase in primary myelodysplastic syndromes: relationship to development of acute myeloid leukaemia. British Journal of Haematology 1995; 90: 125130.
  • 19
    Alvarez FJ, Kisseberth WC, Gallant SL and Couto CG. Dexamethasone, melphalan, actinomycin D, cytosine arabinoside (DMAC) protocol for dogs with relapsed lymphoma. Journal of Veterinary Internal Medicine 2006; 20: 117883.
  • 20
    Dervisis NG, Dominguez PA, Sarbu L, Newman RG, Cadile CD, Swanson CN and Kitchell B. Efficacy of temozolomide or dacarbazine in combination with an anthracycline for rescue chemotherapy in dogs with lymphoma. Journal of the American Veterinary Medical Association 2007; 231: 563569.
  • 21
    Garrett LD, Thamm DH, Chun R, Dudley R and Vail DM. Evaluation of a 6-month chemotherapy protocol with no maintenance therapy for dogs with lymphoma. Journal of Veterinary Internal Medicine 2002; 16: 704709.
  • 22
    Nakamura N, Momoi Y, Watari T, Yoshino T, Tsujimoto H and Hasegawa A. Plasma thymidine kinase activity in dogs with lymphoma and leukemia. Journal of Veterinary Medical Science 1997; 59: 957960.
  • 23
    Chang CM, Schroeder JC, Huang WY, Dunphy CH, Baric RS, Olshan AF, Dorsey K, Dent G, Cerhan J, Lynch C, Rothman N, Cantor N and Blair A. Non-Hodgkin lymphoma (NHL) subtypes defined by common translocations: utility of fluorescence in situ hybridization (FISH) in a case-control study. Leuk Res 2010; 34: 190195.
  • 24
    Fournel-Fleury C, Ponce F, Felman P, Blavier A, Bonnefont C, Chabanne L, Ghernati I and Magnol JP. Canine T-cell lymphomas: a morphological, immunological, and clinical study of 46 new cases. Veterinary Pathology 2002; 39: 92109.
  • 25
    Ponce F, Magnol JP, Ledieu D, Marchal T, Turinelli V, Chalvet-Monfray K and Fournel-Fleury C. Prognostic significance of morphological subtypes in canine malignant lymphomas during chemotherapy. Veterinary Journal 2004; 167: 158166.
  • 26
    Ponce F, Magnol JP, Marchal T, Chabanne L, Ledieu D, Bonnefont C, Felman P and Fournel-Fleury C. High-grade canine T-cell lymphoma/leukemia with plasmacytoid morphology: a clinical pathological study of nine cases. Journal of Veterinary Diagnostic Investigation 2003; 15: 330337.
  • 27
    Ponce F, Marchal T, Magnol JP, Turinelli V, Ledieu D, Bonnefont C, Pastor M, Delignette ML and Fournel-Fleury C. A morphological study of 608 cases of canine malignant lymphoma in France with a focus on comparative similarities between canine and human lymphoma morphology. Veterinary Pathology 2010; 47: 414433.
  • 28
    Valli VE, Vernau W, de Lorimier LP, Graham PS and Moore PF. Canine indolent nodular lymphoma. Veterinary Pathology 2006; 43: 29.
  • 29
    Gion M and Daidone MG. Circulating biomarkers from tumour bulk to tumour machinery: promises and pitfalls. European Journal of Cancer 2004; 40: 26132622.
  • 30
    Michael A, Relph K and Pandha H. Emergence of potential biomarkers of response to anti-angiogenic anti-tumour agents. International Journal of Cancer 2010; 127: 12511258.
  • 31
    Roobol MJ, Haese A and Bjartell A. Tumour markers in prostate cancer III: biomarkers in urine. Acta Oncologica 2011; 50(Suppl. 1): 8589.
  • 32
    Bjartell A, Montironi R, Berney DM and Egevad L. Tumour markers in prostate cancer II: diagnostic and prognostic cellular biomarkers. Acta Oncol 2011; 50(Suppl. 1): 7684.
  • 33
    Bon GG, Kenemans P, Verstraeten R, van Kamp GJ and Hilgers J. Serum tumor marker immunoassays in gynecologic oncology: establishment of reference values. American Journal of Obstetrics & Gynecology 1996; 174(1 Pt 1): 107114.
  • 34
    Eyestone WH. Malignant lymphoma in the dog. Acta Unio Internationalis Contra Cancrum 1952; 8: 536538.
  • 35
    Ma S, Zhang Y, Huang J, Han X, Holford T, Lan Q, Rothman N, Boyle P and Zheng T. Identification of non-Hodgkin's lymphoma prognosis signatures using the CTGDR method. Bioinformatics 2009; 26: 1521.
  • 36
    Gavazza A, Lubas G, Valori E and Gugliucci B. Retrospective survey of malignant lymphoma cases in the dog: clinical, therapeutical and prognostic features. Veterinary Research Communications 2008; 32(Suppl. 1): S291S293.
  • 37
    Yamazaki J, Baba K, Goto-Koshino Y, Setoguchi-Mukai A, Fujino Y, Ohno K and Tsujimoto H. Quantitative assessment of minimal residual disease (MRD) in canine lymphoma by using real-time polymerase chain reaction. Veterinary Immunology and Immunopathology 2008; 126: 321331.