Rabacfosadine for relapsed canine B-cell lymphoma: Efficacy and adverse event profiles of 2 different doses.

Rabacfosadine (RAB), a novel double prodrug of the acyclic nucleotide phosphonate PMEG, preferentially targets neoplastic lymphocytes with reduced off target toxicity. Historical studies have suggested that every 21-day dosing is effective with acceptable toxicity. The purpose of this study was to evaluate RAB's safety and efficacy at 2 different doses every 21 days in dogs with relapsed B-cell lymphoma. Dogs that had failed 1 doxorubicin-based chemotherapy protocol were eligible for inclusion in this prospective trial. Once enrolled, dogs were randomized to receive RAB at either 0.82 mg/kg or 1.0 mg/kg as a 30-minute IV infusion every 21 days for up to 5 treatments. Response assessment and adverse event (AE) evaluation were performed every 21 days via VCOG criteria. Fifty dogs were enrolled, with 16 treated at 0.82 mg/kg and 34 treated at 1.0 mg/kg. The overall response rate was 74%, with 45% of dogs experiencing a complete response (CR). The median progression free intervals (PFIs) were 108 days, 172 days and 203 days for all dogs, all responders, and all CRs, respectively. Response rates and PFIs were similar in both treatment groups. The incidence of AEs, dose delays, dose reductions and withdrawals were not statistically different between the 2 groups. The AEs observed were similar to those previously reported and included hematologic, gastrointestinal, dermatologic and pulmonary AEs. One dog had grade 5 pulmonary fibrosis; otherwise, AEs resolved with supportive treatment. Rabacfosadine is a generally well tolerated, effective chemotherapy option for dogs with relapsed B-cell lymphoma.

c-Kit Mutation and Localization Status as Response Predictors in Mast Cell Tumors in Dogs Treated with Prednisone and Toceranib or Vinblastine.

BACKGROUND: KIT inhibitors, such as toceranib (TOC), and vinblastine (VBL) have not been prospectively compared in the treatment of macroscopic mast cell tumors (MCTs). Also, it is unknown whether VBL or TOC is superior for treating MCT without c-kit mutations. HYPOTHESIS/OBJECTIVES: To determine the value of KIT genotyping and localization in treatment decisions for dogs with macroscopic MCT. We hypothesized that c-kit mutated MCT would have a better response to TOC than VBL. ANIMALS: Eighty-eight client-owned dogs with macroscopic MCT. METHODS: Prospective, randomized trial. Dogs were randomized to TOC (2.75 mg/kg EOD) or VBL (2.5 mg/m2 weekly × 4 then EOW) by KIT localization and c-kit mutation status using an adaptive randomization scheme. RESULTS: Sixty dogs were allocated to TOC and 28 to VBL. Of the dogs receiving TOC, 20% had c-kit mutations, compared to 30% receiving VBL (P = 0.74). Overall response rates were 46% (TOC) and 30% (VBL) (odds ratio = 1.56 [0.62-3.92]; P = 0.28). Median progression-free survival (PFS) for dogs receiving VBL was 78 days (7-1,521) and for TOC 95.5 (14-990); hazard ratio (HR) = 1.34 [0.72-2.50]; P = 0.36. Median overall survival (OS) was 241.5 days (10-1,521) for the VBL group and 159 (20-990) for the TOC group; HR = 0.80 ([0.45-1.41]; P = 0.44). CONCLUSIONS AND CLINICAL IMPORTANCE: Neither PFS nor OS was significantly different between treatment groups. As the proportion of dogs with c-kit mutations was not different between treatment groups in this population of dogs, c-kit mutation status did not predict treatment response.

Extravasation reactions associated with the administration of pamidronate: 11 cases (2008-2013).

Pamidronate is a bisphosphonate drug widely utilized in veterinary oncologic practice for the palliation of malignant osteolysis. Pamidronate has not been previously reported to cause tissue injury upon extravasation in dogs. The medical records of 11 client-owned dogs undergoing palliative treatment for primary bone tumors with known or suspected pamidronate extravasation reactions were reviewed. The majority of adverse events were low grade in nature, however in some cases, the reactions were severe and led to euthanasia in one instance. Time to complete resolution of lesions ranged from within several days to greater than one and a half months. Aside from the dog that was euthanized, no long-term sequelae of extravasation were identified. Treatments employed to address the reactions varied widely. Pamidronate extravasation reaction appears to be an uncommon, but potentially serious complication of intravenous administration.

Doxorubicin chemotherapy for presumptive cardiac hemangiosarcoma in dogs†.

Sixty-four dogs were treated with single-agent doxorubicin (DOX) for presumptive cardiac hemangiosarcoma (cHSA). The objective response rate (CR + PR) was 41%, and the biologic response rate (CR + PR + SD), or clinical benefit, was 68%. The median progression-free survival (PFS) for treated dogs was 66 days. The median survival time (MST) for this group was 116 days and was significantly improved compared to a MST of 12 days for untreated control dogs (P = 0.0001). Biologic response was significantly associated with improved PFS (P < 0.0001) and OS (P < 0.0001). Univariate analysis identified larger tumour size as a variable negatively associated with PFS. The high rate of clinical benefit and improved MST suggest that DOX has activity in canine cHSA.

Pulse-Administered Toceranib Phosphate Plus Lomustine for Treatment of Unresectable Mast Cell Tumors in Dogs.

BACKGROUND: Nonresectable mast cell tumors (MCT) in dogs remain a therapeutic challenge, and investigation of novel combination therapies is warranted. Intermittent administration of tyrosine kinase inhibitors (TKI) combined with cytotoxic chemotherapy may effectively chemosensitize canine MCT while decreasing cost and adverse effects associated with either agent administered as monotherapy. HYPOTHESIS/OBJECTIVES: The primary study objectives were to (1) identify the maximally tolerated dose (MTD), (2) determine the objective response rate (ORR) and (3) describe the adverse event profile of pulse-administered toceranib phosphate (TOC) combined with lomustine. ANIMALS: Forty-seven client-owned dogs with measurable MCT. METHODS: Toceranib phosphate was given PO on days 1, 3 and 5 of a 21-day cycle at a target dosage of 2.75 mg/kg. Lomustine was given PO on day 3 of each cycle at a starting dosage of 50 mg/m(2) . All dogs were concurrently treated with diphenhydramine, omeprazole, and prednisone. RESULTS: The MTD of lomustine was established at 50 mg/m(2) when combined with pulse-administered TOC; the dose-limiting toxicity was neutropenia. Forty-one dogs treated at the MTD were evaluable for outcome assessment. The ORR was 46% (4 complete response, 15 partial response) and the overall median progression-free survival (PFS) was 53 days (1 to >752 days). On multivariate analysis, variables significantly associated with improved PFS included response to treatment, absence of metastasis, and no previous chemotherapy. CONCLUSIONS AND CLINICAL IMPORTANCE: Combined treatment with pulse-administered TOC and lomustine generally is well tolerated and may be a reasonable treatment option for dogs with unresectable or metastatic MCT.

Early tumor response to intraarterial or intravenous administration of carboplatin to treat naturally occurring lower urinary tract carcinoma in dogs.

BACKGROUND: Survival times and tumor responses associated with malignant neoplasia of the lower urinary tract are poor despite the vast array of current treatments. Therefore, the evaluation of alternative treatments, such as intraarterial administration of chemotherapy (IAC) should be considered. OBJECTIVE: To describe a technique for superselective catheterization for IAC and to evaluate initial tumor response by ultrasonography after both IAC and intravenous administration of chemotherapy (IVC). ANIMALS: Client-owned dogs with lower urinary tract neoplasia treated with either IVC (n = 15) or IAC (n = 11). METHODS: Retrospective study. An arterial approach via the carotid or femoral artery was utilized to obtain superselective access and administer chemotherapy in the IAC cases. Medical record review was performed, data were recorded, and recorded variables were evaluated statistically. RESULTS: Intraarterial chemotherapy was successfully administered in all cases. There was a significantly greater decrease in longest unidimensional measurement in the IAC group as compared to the IVC group (P = .013). The IAC group was also significantly more likely to have a tumor response as assessed by modified RECIST guidelines (P = .049). Dogs in the IAC group were significantly less likely to develop anemia (P = .001), lethargy (P = .010) and anorexia (P = .024). CONCLUSION AND CLINICAL IMPORTANCE: This study demonstrated the feasibility and efficacy of performing IAC for lower urinary tract neoplasia. Further investigation is necessary as the follow-up time was short and the impact on long-term outcome and survival was not determined.

Feline exocrine pancreatic carcinoma: a retrospective study of 34 cases.

Thirty-four cases were reviewed in this retrospective study for information on clinical presentation, prognostic indicators, survival time and response to various therapies. The most common presenting clinical signs were weight loss, decreased appetite, vomiting, palpable abdominal mass and diarrhoea. Metastatic disease was confirmed in 11 cats. The overall median survival was 97 days. The median survival times for patients who received chemotherapy or had their masses surgically removed was 165 days. Those patients who had an abdominal effusion present at the time of diagnosis survived a median of 30 days. Cats that received non-steroidal anti-inflammatory drug therapy had a median survival of 26 days. This study confirms that exocrine pancreatic carcinoma in cats is an aggressive tumour with a high metastatic rate and poor prognosis, although three patients survived over 1 year. Fifteen percent of the patients were diabetic, which raises the question as to what the link between diabetes and pancreatic cancer in people and cats may be.

Efficacy of doxorubicin-based chemotherapy for non-resectable canine subcutaneous haemangiosarcoma.

Eighteen dogs with measurable subcutaneous haemangiosarcoma (SQHSA) were treated with doxorubicin-based chemotherapy. Response assessment was evaluated and compared using World Health Organization (WHO), Response Evaluation Criteria in Solid Tumours (RECIST) and tumour volume criteria. The overall response rate for all dogs was 38.8% using WHO criteria, 38.8% using RECIST criteria and 44% using tumour volume criteria. One dog had a complete response. The median response duration for all dogs was 53 days (range 13-190 days). Four dogs had complete surgical excision after neoadjuvant chemotherapy. The median progression-free interval for dogs with complete surgical excision after neoadjuvant chemotherapy was significantly longer than those not having surgical excision (207 days versus 83 days, respectively) (P = 0.003). No significant difference in metastasis-free interval or survival time was found between the groups. Doxorubicin-based chemotherapy appears to be effective for non-resectable canine SQHSA, although the response duration is relatively short.

Long-term survival in dogs with localized histiocytic sarcoma treated with CCNU as an adjuvant to local therapy.

Histiocytic sarcoma (HS) is associated with a poor prognosis owing to the presence of metastasis at the time of diagnosis in most dogs. Improved outcome has been reported in several dogs with localized HS following local therapy, however, distant metastasis occurs in 70-91% of dogs suggesting that adjuvant systemic therapy is necessary. The purpose of this retrospective study was to describe clinical characteristics and outcome in dogs with localized HS treated with aggressive local therapy plus adjuvant CCNU chemotherapy. Data from 16 dogs were evaluated. The median disease-free interval was 243 days. Two dogs had local recurrence and eight dogs developed metastatic disease with a median time to relapse of 201 days in these 10 dogs. The median survival time for all 16 dogs was 568 days. These results support the recommendation for aggressive local therapy combined with adjuvant CCNU chemotherapy in dogs with localized HS.

Survival analysis of 97 cats with nasal lymphoma: a multi-institutional retrospective study (1986-2006).

BACKGROUND: Feline nasal lymphoma (NLSA) is a condition for which no standard of care exists. HYPOTHESIS: There is no difference in survival times of cats with NLSA treated with single or multimodality therapy. ANIMALS: Records from 97 cats diagnosed with NLSA were examined. METHODS: The purpose of this retrospective study was to compare the survival times of cats with NLSA treated with radiation therapy (RT) alone, chemotherapy alone, or RT + chemotherapy and identify potential prognostic variables that affected survival. Cats were grouped according to therapy: RT + chemotherapy (n = 60), RT alone (n = 19), or chemotherapy alone (n = 18). RESULTS: Survival was calculated with 2 methods. The 1st survival analysis (method A) included all cats, but counted only deaths caused by progressive NLSA. The median survival time (MST), regardless of therapy modality, was 536 days. The 2nd survival analysis (method B) also included all cats and counted all deaths, regardless of cause, as events. The overall MST calculated for all deaths was 172 days. A negative independent prognostic variable identified was anemia (P < .001), and positive independent prognostic variables were a complete response to therapy (P < .001) and total radiation dose >32 Gy (P= .03). CONCLUSIONS AND CLINICAL IMPORTANCE: There were no significant differences in survival times among the 3 treatment groups but these results suggest that the addition of higher doses of RT to a cat's treatment protocol may control local disease and therefore influence survival.

ABCB1-1Delta polymorphism can predict hematologic toxicity in dogs treated with vincristine.

BACKGROUND: Dogs that harbor the naturally occurring ABCB1-1Delta polymorphism experience increased susceptibility to avermectin-induced neurological toxicosis as a result of deficient P-glycoprotein function. Whether or not the ABCB1-1Delta polymorphism affects susceptibility to toxicity of other P-glycoprotein substrate drugs has not been studied. HYPOTHESIS: Dogs that possess the ABCB1-1Delta mutation are more likely to develop hematologic toxicity associated with vincristine than ABCB1 wild-type dogs. ANIMALS: Thirty-four dogs diagnosed with lymphoma were included in this study. METHODS: Cheek swab samples were obtained from dogs diagnosed with lymphoma that were to be treated with vincristine. DNA was extracted from cheek swabs and the ABCB1 genotype was determined. Hematologic adverse drug reactions were recorded for each dog and graded according to the Veterinary Comparative Oncology Group's criteria for adverse event reporting (Consensus Document). In order to avoid possible bias, ABCB1 genotype results for a particular patient were not disclosed to oncologists until an initial adverse event report had been submitted. RESULTS: Dogs heterozygous or homozygous for the ABCB1-1Delta mutation were significantly more likely to develop hematologic toxicity, specifically neutropenia (P= .0005) and thrombocytopenia (P= .0001), after treatment with vincristine than ABCB1 wild-type dogs. CONCLUSIONS AND CLINICAL IMPLICATIONS: At currently recommended dosages (0.5-0.7 mg/M(2)), vincristine is likely to cause hematologic toxicity in dogs with the ABCB1-1Delta mutation, resulting in treatment delays and unacceptable morbidity and mortality. Assessing the ABCB1-1Delta genotype before vincristine administration and decreasing the dosage may prevent toxicity and treatment delays resulting from neutropenia or thrombocytopenia.

RNA-loaded CD40-activated B cells stimulate antigen-specific T-cell responses in dogs with spontaneous lymphoma.

Cell-based vaccination strategies to induce functional tumor-specific T cells in cancer patients have focused on using autologous dendritic cells. An alternative approach is to use RNA-loaded CD40 activated B cells (CD40-B) that are highly efficient antigen-presenting cells capable of priming naive T cells, boosting memory T-cell responses and breaking tolerance to tumor antigens. The use of tumor RNA as the antigenic payload allows for gene transfer without viruses or vectors and permits major histocompatibility complex (MHC)-independent, multiple-antigen targeting. Here, we use CD40L transfected K562 cells to generate functional CD40-B cells from the peripheral blood of humans and dogs. Testing of RNA-loaded CD40-B cells in dogs allows not only for its development in veterinary medicine but also for determination of its safety and efficacy in a large animal model of spontaneous cancer prior to initiation of human clinical trials. We found that CD40-B cells from healthy humans, healthy dogs and tumor-bearing dogs express increased levels of immune molecules such as MHC and CCR7. Moreover, RNA-loaded CD40-B cells induce functional, antigen-specific T cells from healthy dogs and dogs with lymphoma. These findings pave the way for immunotherapy trials using tumor RNA-loaded CD40-B cells to stimulate antitumor immunity in a large animal model of spontaneous neoplasia.

Description of clinical and pathological findings, treatment and outcome of feline large granular lymphocyte lymphoma (1996-2004).

Feline large granular lymphocyte (LGL) lymphoma is an uncommon, morphologically distinct variant of feline lymphoma. Limited information exists in the literature regarding pathological and immunohistochemical descriptions, clinical findings, treatment and survival times. The purpose of this study was to describe clinical features, treatment and outcome in feline LGL lymphoma. Medical records of 45 cats with LGL lymphoma were retrospectively evaluated. Decreased appetite/anorexia, weight loss, lethargy and vomiting were the most commonly reported clinical signs. All cats tested for feline leukaemia virus and feline immunodeficiency virus infection were negative. The mesenteric lymph nodes and small intestine were the most commonly affected organs. One complete response and six partial responses were noted in the 23 cats that received chemotherapy as their initial treatment. Median survival time for cats that were treated was 57 days. Based on these results, feline LGL lymphoma appears to be minimally responsive to chemotherapy and is associated with a grave prognosis.

Characterization of the biological behaviour of appendicular osteosarcoma in Rottweilers and a comparison with other breeds: a review of 258 dogs.

The purpose of this retrospective study was to compare Rottweilers diagnosed with osteosarcoma (OSA) with other breeds to determine whether Rottweilers experienced a more aggressive form of the disease. Two hundred and fifty-eight dogs were evaluated (102 clinical and 156 necropsy cases). In the necropsy population, Rottweilers had a younger mean age at death (7.3 versus 9 years, P = 0.006). There were no significant differences between Rottweilers and other breeds in age at diagnosis, median disease-free interval or survival time. However, Rottweilers were more likely to have metastasis to the brain (7 versus 0%, P = 0.03). These results suggest that OSA in Rottweilers may have a different biological behaviour, but this study did not confirm that these differences were associated with a worse outcome.

Response of canine cutaneous epitheliotropic lymphoma to lomustine (CCNU): a retrospective study of 46 cases (1999-2004).

BACKGROUND: Epitheliotropic lymphoma (ELSA) is an uncommon cutaneous canine malignancy of T lymphocytes. A consensus regarding the therapeutic standard of care is lacking, warranting evaluation of chemotherapeutic agents traditionally employed against canine nodal lymphoma in the treatment of ELSA. HYPOTHESIS: The purpose of this retrospective, multi-institutional study was to evaluate the efficacy of 1-(2-chloroethyl)-3-cyclohexyl-l-nitrosourea (CCNU) in the treatment of ELSA. ANIMALS: Forty-six dogs with adequate follow-up and treatment response information. METHODS: All cases were diagnosed histopathologically. Immunohistochemisty (CD3, CD79a) was performed on 42/46 samples. RESULTS: Presenting skin lesions included generalized scales (25/46), plaques or nodules (22/46), mucocutaneous lesions (14/ 46), and corneal involvement (1/46). Lymph node involvement and Sézary syndrome were documented in 7 and 2 dogs, respectively. The median number of CCNU treatments was 4 (range, 1-11), with a median starting dose of 60 mg/m(2) (range, 30-95). Of the 46 dogs, 15 achieved complete remission, 23 achieved partial remission, 5 had stable disease, and 3 had progressive disease, for an overall response rate of 83%. The median number of treatments to achieve a response was 1 (range, 1-6). The overall median duration of response was 94 days (range, 22-282). Sixteen dose reductions were required because of neutropenia (10/46), thrombocytopenia (1/46), anemia (1/46), increased liver enzyme activity (3/46), or unspecified reasons (1/46). CONCLUSIONS AND CLINICAL IMPLICATIONS: Given the high response rate and well tolerated protocol, prospective studies are warranted to investigate the utility of CCNU alone or in multi-agent protocols for the treatment of ELSA.

Cyclooxygenase-2 expression is associated with histologic tumor type in canine mammary carcinoma.

Cyclooxygenase-2 (COX-2) is an inducible member of the family of cyclooxygenase enzymes that has been implicated in the genesis of numerous cancers. The role of COX-2 in canine mammary neoplasia remains to be more clearly elucidated. The goal of the study reported here was to determine whether a direct association between levels of COX-2 expression and tumor histologic subtype exists in canine mammary carcinoma. Immunohistochemical analysis was performed using a polyclonal antiprostaglandin G/H synthase 2 IgG COX-2 antibody. Sections from the kidneys of young dogs, which stain positive for COX-2 in the macula densa, served as positive controls. Slides were reviewed by a single pathologist, and were evaluated for COX-2 expression according to previously established scales. Positive-staining tumors were given a COX-2 staining distribution (on the basis of the percentage of positive staining cells in five 400x fields) and intensity score according to previously established scales. The product of the COX-2 staining distribution and intensity scores was calculated to create a COX-2 staining index. COX-2 expression was detected in 28 of 50 (56%) samples evaluated. Anaplastic carcinomas had a significantly higher COX-2 staining distribution, intensity, and index, compared with those for adenocarcinomas (P < 0.0001). The overall percentage of positive tumors (56%) was consistent with that of prior studies. To the authors' knowledge, these results indicate, for the first time, a direct association between COX-2 expression and tumor histologic subtype in canine mammary carcinomas. Future research directed at measuring tumor response in canine mammary carcinoma patients treated with a selective COX-2 inhibitor is indicated.

Assessment of cyclooxygenase-2 expression in canine hemangiosarcoma, histiocytic sarcoma, and mast cell tumor.

To determine whether cyclooxygenase-2 (COX-2) is expressed in canine hemangiosarcoma (HSA), histiocytic sarcoma (HS), and grade-II mast cell tumor (MCT), we performed immunohistochemistry using COX-2 antibodies in the aforementioned tumors. Twenty cases of each tumor type were selected initially from the Laboratory of Pathology archives of cases submitted through the Matthew J. Ryan Veterinary Hospital of the University of Pennsylvania. Immunohistochemistry was performed, using a polyclonal antiprostaglandin endoperoxide synthase immunoglobulin G COX-2 antibody. Sections from the kidneys of young dogs, in which the macula densa stains positive for COX-2, served as positive controls. Slides were reviewed by a single pathologist (M. H. Goldschmidt) and graded for COX-2 expression according to previously established scales. Descriptive data is given for each tumor type. COX-2 expression was identified in 0 of 19 HSA, 1 of 20 HS, and 1 of 17 grade-II MCT. Although COX-2 has been shown to be overexpressed in selected human sarcomas and hematopoeitic tumors, these results indicate that canine HSA, HS, and MCT do not express COX-2 in any appreciable fashion.

Progression of a solitary, malignant cutaneous plasma-cell tumour to multiple myeloma in a cat.

An 11-year-old male domestic shorthair cat was examined because of a soft-tissue mass on the left tarsus previously diagnosed as a malignant extramedullary plasmacytoma. Findings of further diagnostic tests carried out to evaluate the patient for multiple myeloma were negative. Five months later, the cat developed clinical evidence of multiple myeloma based on positive Bence Jones proteinuria, monoclonal gammopathy and circulating atypical plasma cells. This case represents an unusual presentation for this disease and documents progression of an extramedullary plasmacytoma to multiple myeloma in the cat.

Cardiac troponin I in the normal dog and cat.

Cardiac troponin I (cTnI) has proven to be a highly specific and sensitive marker for myocardial cellular damage in many mammalian species. The structure of cTnI is highly conserved across species, and assays for human cTnI (including the one used in the current study) have been validated in the dog. Blood concentrations of cTnI rise rapidly after cardiomyocyte damage, and assay of cTnI potentially may be valuable in many clinical diseases. The purpose of this study was to establish the normal range of cTnI in heparinized plasma of dogs and cats. Forty one clinically normal dogs and 21 cats were included in the study. One to 3 milliliters of blood were collected by venipuncture into lithium heparin vacutainers for analysis of cTnI (Stratusz CS). The range of plasma cTnI concentrations in dogs was <0.03 to 0.07 ng/mL with a mean of 0.02 ng/mL, with the upper tolerance limit (0.07 ng/mL) at the 90th percentile with 95% confidence. In cats, the range was <0.03 to 0.16 ng/mL with a mean of 0.04 ng/mL, and the upper tolerance limit (0.16 ng/mL) at the 90th percentile as well with 90% confidence. This study establishes preliminary normal ranges of plasma cTnI in normal dogs and cats for comparison to dogs and cats with myocardial injury or disease.

Plasma vascular endothelial growth factor concentrations in healthy dogs and dogs with hemangiosarcoma.

Vascular endothelial growth factor (VEGF) is a dimeric glycosylated polypeptide growth factor with potent angiogenic, mitogenic, and vascular permeability-enhancing properties specific for endothelial cells. In humans, VEGF seems to play a major role in tumor growth, and plasma concentrations correlate with tumor burden, response to therapy, and disease progression. This study compared plasma VEGF concentrations in healthy client-owned dogs (n = 17) to dogs with hemangiosarcoma (HSA; n 16). Dogs with HSA were significantly more likely to have detectable concentrations of plasma VEGF (13/17) compared to healthy dogs (1/17; P < .001). The median plasma VEGF concentration for dogs with HSA was 17.2 pg/mL (range, < 1.0-66.7 pg/mL). Plasma VEGF concentrations in dogs with HSA did not correlate with stage of disease or tumor burden, but 1 dog had undetectable VEGF during chemotherapy that subsequently increased with disease progression.