Adjuvant CCNU (Lomustine) and Prednisone Chemotherapy for Dogs With Incompletely Excised Grade 2 Mast Cell Tumors

Introduction

Mast cell tumor (MCT) is the most common malignant cutaneous/ subcutaneous tumor in dogs, accounting for 11% to 27% of all malignant cutaneous tumors.¹ The primary treatment modality for grade 1 or 2 MCTs is wide (2 to 3 cm) surgical excision. For dogs in which complete surgical excision cannot be achieved, adjuvant radiation therapy has been used with success; local recurrence rates of 5% to 20% have been reported.^2–6 However, according to these reports, 11% to 31% of dogs have multiple MCTs at the time of diagnosis; 10% to 31% of cases develop second primary cutaneous MCTs during the follow-up period; and 0% to 12% of cases develop metastases. Such results indicate that aggressive local radiation therapy is less likely to achieve a cure.^2–6

Mast cell tumors also have moderate sensitivity to chemotherapy;^7–11 vinblastine, prednisone, and 1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea (CCNU; lomustine) have been reported to be active against MCTs.^7,8,11 The reported response rate is 20% with prednisone alone, 47% with a combination of vinblastine and prednisone, and 42% with single-agent CCNU.^7,8,11 The reported response rates for single-agent vinblastine range from 12%¹² to 27%.¹³

A combination of vinblastine and prednisone has been evaluated as an alternative to radiation therapy for dogs with microscopic residual MCTs after surgery. One study reported local recurrence rates of 10%, and the other reported 1- and 2-year disease-free rates of 57%.^11,14 However, the former study also included MCTs with complete but close margins, posing potential overestimation of the treatment efficacy.¹⁴ Histological determination of a tumor margin is based on only a small portion of the surgical specimen, and distinguishing between incompletely and completely excised tumors (as opposed to the viewed margin) is not always possible and sometimes can even be misleading. To the authors’ knowledge, CCNU has not been evaluated in dogs with microscopic residual MCTs after surgery.

The purpose of this study was to retrospectively evaluate the efficacy of adjuvant CCNU, with or without concurrent prednisone, in dogs with incompletely excised grade 2 MCTs. The authors hypothesized that adjuvant CCNU/prednisone chemotherapy provides a local control rate equivalent to that of radiation therapy and is a feasible alternative treatment for incompletely excised grade 2 MCTs in dogs.

Materials and Methods

The medical records database (1994 through 2004) was searched for dogs with MCTs that were treated with surgery and chemotherapy. The inclusion criteria were 1) histopathological diagnosis of grade 2 MCT with or without metastases, or grade 1 MCT with regional metastases; 2) incomplete surgical excision of the gross tumor volume (primary and metastatic lymph nodes); and 3) use of post-operative CCNU chemotherapy (with or without prednisone), started within 2 months of the surgery.

Dogs with grade 3MCTs, with or without metas-tases, and dogs with grade 1 MCTs without metastases were excluded from the study. Tumors that histologically extended “close” to the surgical margins (i.e., <1 mm) were considered as having complete margins in order to exclude “false incomplete excision,” and only tumors that histologically extended to the actual surgical margin were included in the study. Since disagreement is considerable in the grading of MCTs between pathologists,¹⁵ all available surgical specimens were reviewed by two pathologists (Beamer/Stromberg) to evaluate the grade¹⁶ and completeness of the excision.

The following information was retrieved from the medical records: age, breed, gender, body weight, initial tumor size (longest diameter), number of MCTs, previous history of MCTs, interval between surgery and initiation of CCNU chemotherapy, the number of doses and dosage of CCNU, the duration of CCNU therapy, and the dose of prednisone if it was prescribed. Hematological toxicities were assessed based on complete blood counts (CBCs) obtained immediately before and 6 to 8 days after each dose of CCNU. Biochemical toxicities were assessed based on the serum biochemical profile values immediately prior to each dose of CCNU. Gastrointestinal (GI) toxicities were assessed based on the owner’s description of the clinical signs and the description in the medical record. Toxicity was graded according to the Veterinary Cooperative Oncology Group Common Terminology for Adverse Events (VCOGCTCAE) v. 1.0¹⁷ [see Table]. Hepatotoxicity was monitored by measuring serum alanine transaminase (ALT) activity and was graded according to the VCOG-CTCAE v. 1.0 (reference range 10 to 55 IU/L; grade 1, 56 to 82 IU/L; grade 2, 83 to 110 IU/L; grade 3, 111 to 550 IU/L; grade 4, >550 IU/L).

Local recurrence was defined as MCT development within 3 cm of the initial surgical site. A single cutaneous MCT development outside this area was considered a second primary tumor and was distinguished from distant metastases. Regional metastasis was defined as histological or cytological evidence of MCT in the regional lymph node. Distant metastasis was defined as MCT development in noncutaneous/subcutaneous sites or development of multiple simultaneous cutaneous MCTs.

The endpoints for progression-free interval (PFI) were development of local recurrence, regional or distant metastases, and treatment-related death. Dogs lost to follow-up, dead from an unrelated cause, or alive at the time of data analysis were censored from PFI analysis. The endpoints for progression-free survival (PFS) were development of local recurrence, regional or distant metastases, or death from any cause. Dogs lost to follow-up and alive at the time of data analysis were censored from PFS analysis. Survival curves for PFI and PFS were generated using the Kaplan-Meier product limit method and commercial computer software.^a

Results

Twelve dogs met the inclusion criteria. Nine were female (all spayed) and three were male (one intact). The breeds represented included three cocker spaniels, three Labrador retrievers, three mixed-breed dogs, two boxers, and one English pointer. The median age was 9 years (range 3 to 13 years). The median body weight was 22.9 kg (range 11.0 to 39.7 kg).

Initial histopathological grade was grade 2 in 11 dogs and grade 1 in one dog; the latter had a metastasis to the regional lymph node. Specimens from the initial tumor were available in 10 dogs. After reviewing these histopathological specimens, all MCTs (including the one previously diagnosed as grade 1) were classified as grade 2. At the time of surgery, two dogs had evidence of regional metastases (surgically removed), and four dogs had more than one MCT. The median tumor size was 2.5 cm (range 0.5 to 7.0 cm). Four MCTs were located in the trunk/flank, four MCTs were in the inguinal/perineal area, four MCTs were in the extremity, and one MCT was in the neck.

The median time from the surgical excision to the initiation of CCNU chemotherapy was 14.5 days (range 9 to 47 days). The median initial dosage of CCNU was 60 mg/m² q 3 weeks (range 56.7 to 83.9 mg/m²). Dose reduction (n=2), interval prolongation (n=3), or both (n=1) were made due to neutropenia (n=2), increased serum ALT activity (n=2), or clinician’s preference (n=2) in subsequent CCNU doses, resulting in a median dose intensity of 17 mg/m² per week (range 9 to 24 mg/m² per week). The dog that received the lowest dose intensity was the one with the lowest body weight (11.0 kg). Both dose reduction and interval prolongation were made in this dog according to the clinician’s decision, although the dog did not develop significant toxicity after the first dose of CCNU. Prednisone was concurrently used in all but one dog, at a dosage of 40 mg/m² q 24 hours for the first 7 days, and then 20 mg/m² q 48 hours thereafter.

All dogs were intended to receive chemotherapy for at least 4 to 6 months or until an unacceptable toxicity had occurred. The median duration of CCNU treatment was 6.3 months (range 1.4 to 15.2 months), and the median number of CCNU doses was eight (range three to 22 doses). The intended treatment was completed in nine dogs; CCNU was discontinued due to increased serum ALT activity in three dogs after three, three, and six doses of CCNU, respectively. The median cumulative dose was 400 mg/m² (range 178 to 1615 mg/m²).

The median follow-up time was 620 days (range 184 to 1384 days). At the time of analysis, two dogs were still alive at 1032 and 1384 days, respectively; four dogs were lost to follow-up at 254, 512, 712, and 763 days, respectively; two dogs died of unrelated causes at 184 (osteosarcoma) and 886 days (chronic renal failure), respectively; two dogs died of suspected CCNU-induced hepatopathy at 500 and 568 days, respectively; and two dogs died of an unknown cause at 515 and 672 days, respectively. None of the dogs developed local recurrence or regional/distant metastases. One dog (8%) developed distant cutaneous MCTs at 618 days (403 days after the last CCNU dose). Three additional dogs developed a second malignant neoplasm; one developed an oral melanoma at day 662 (317 days after the last CCNU dose), one developed a thyroid carcinoma at day 1363 (1147 days after the last CCNU dose), and the other developed an appendicular osteosarcoma at day 184 (2 days after the last CCNU dose). The Kaplan-Meier curve for PFI is shown in the Figure. The 1- and 2-year PFI rates were 100% and 77%, respectively (development of a “de novo” MCT was not included). The median PFI was not reached; the median PFS was 886 days.

Complete blood count 7 days after the first CCNU dose (the typical nadir for neutropenia after CCNU) was available in seven dogs. Neutropenia considered to be grade 3 or higher was seen in two dogs (29%; one was grade 3 after 57.6 mg/m² of CCNU, and one was grade 4 after 83.9 mg/m² of CCNU). No dogs developed anemia or thrombocytopenia. Data on GI toxicity were available in all 12 dogs. None of the dogs experienced GI toxicity that was grade 3 or higher; grade 2 diarrhea was reported in three dogs, and one of those dogs had concurrent grade 1 vomiting and grade 1 anorexia/lethargy.

Serial serum biochemical profiles during CCNU therapy were available in nine of 12 dogs. Serum biochemical profiles were not initially monitored in three dogs. Increased serum ALT activities that were grade 3 or higher were seen in nine (75%) of the 12 dogs; one dog developed grade 4 hepatotoxicity, and two additional dogs died of hepatic disease. The serum ALT activities in these dogs were 521 U/L and 444 U/L, respectively, prior to developing liver failure. In both cases, nonspecific, multifocal hepatocellular necrosis was confirmed by histopathology of percutaneous biopsy samples. The first dog developed liver failure 315 days after receiving 10 doses of CCNU at a dose intensity of 19.3 mg/m² per week (cumulative dose 579 mg/m²), and the second dog developed liver failure at 500 days after receiving 22 doses of CCNU at a dose intensity of 23.7 mg/m² per week (cumulative dose 1615 mg/m²). For these two dogs, the intervals between the last dose of CCNU and the development of clinical signs of liver failure were 246 days and 49 days, respectively.

Discussion

Local fractionated radiation therapy is a widely accepted recommendation for successful treatment of incompletely excised grade 1 and 2 MCTs in dogs that are not eligible for a second surgery.¹⁸ Although excellent outcomes have been reported with radiation therapy, its true necessity is currently being questioned. A recent study revealed only a 23% local recurrence rate after incomplete excision and oral administration of prednisone alone, although higher recurrence rates have been reported in dogs not receiving prednisone.^19–21

Compared to chemotherapy, fractionated radiation therapy has several disadvantages: 1) limited accessibility of radiation facilities in veterinary medicine precludes this option for some; 2) acute skin reactions prevent this therapy from being ideal in locations such as the face or perineal area; and 3) it is expensive and labor intensive, especially for treatment of multiple sites. This study excluded dogs with grade 3 MCTs in order to have a case population comparable with that of previous studies evaluating adjuvant radiation therapy; however, chemotherapy may also be effective for dogs with grade 3 MCTs, given the high risk of recurrence and metastases. Local recurrence or regional/ distant metastases were not seen in any of the dogs in this study, suggesting that CCNU may be an alternative treatment option to radiation therapy. This conclusion could not be made, however, as the sample size of this study was small and the lengths of treatments were inconsistent. Further evaluation with a larger population of dogs is warranted.

Development of another MCT during the follow-up period is a major problem that makes an expensive local treatment less feasible for dogs with histories of multiple MCTs. Whether these subsequent cutaneous MCTs are clonal or not is currently unknown. Clonality of multiple subsequent MCTs developing in the same dog has been demonstrated in two dogs.²² Speculation is that the development of new MCTs may be prevented by chemotherapy if they are clonal in origin and are present microscopically at the time of chemotherapy. In the current study, development of a new single, cutaneous MCT was seen in one dog (8%)—a finding similar to that of previous studies.^23–25 The new MCT was considered a “de novo” tumor rather than a distant metastasis, because evidence was insufficient to define this lesion as metastatic.

Hematological and GI toxicities were infrequent, and the treatment protocol was generally well tolerated, although neutropenia may have been determined to be more prevalent if CBCs had been obtained 7 days after the CCNU doses in all cases. Based on this limited study, a major complication of CCNU treatment is hepatotoxicity. Increases in serum ALT activities grade 3 or higher were observed in 75% of the dogs, although concurrent corticosteroid administration possibly contributed to the observed ALT elevation. Fatal liver failures were observed in a relatively high proportion (17%) of the dogs; however, those dogs that died received high cumulative doses of CCNU without having their serum biochemical profiles monitored (an established monitoring protocol was lacking early in the study). No dog that had its serum ALT activity monitored closely developed liver failure.

Hepatotoxicity is a well-known, although poorly understood, adverse effect of CCNU administration. A high cumulative dose of CCNU is associated with development of liver failure.^26–28 No guideline currently exists for monitoring for this potentially fatal toxicity or for adjusting the dosages of CCNU according to the changes in biochemical parameters. At the authors’ practice, a discontinuation of CCNU therapy or a dose/interval adjustment is recommended if the serum ALT activity exceeds 250 IU/L; also, treatment length should not be >6 months. Further investigation of CCNU-related hepatotoxicity is warranted, which should assist in providing more definitive guidelines.

Conclusion

Lomustine chemotherapy is a reasonable treatment alternative to radiation therapy for incompletely excised grade 2 MCTs in dogs. Further studies with a larger population and a fixed treatment and monitoring protocol are warranted.