J Vet Intern Med 2006;20:1398–1401
Hematologic Changes Associated with Half-Body Irradiation in Dogs with Lymphoma Sandra M. Axiak, Janet K. Carreras, Kevin A. Hahn, Melissa M. Endicott, Dorothy E. Parshley, and Glen K. King Background: Reports describe the technique and efficacy of half-body irradiation (HBI) of dogs with lymphoma, but few describe the distinctive toxicoses associated with the combination of HBI and chemotherapy. Hypothesis: HBI would transiently affect myelocytic and erythroid variables as assessed by serial analysis of complete blood counts. Animals: Twenty-nine dogs with lymphoma treated with HBI during 2002 and 2003. Methods: A retrospective study of medical records of 29 dogs was performed. Two HBI protocols were used, resulting in delivery of either 6 Gy or 8 Gy to each half of the body, 1 month apart. Dogs received chemotherapy before, during, or after irradiation, or at multiple times. Serial hematology was available for all dogs. Data were analyzed between collection periods by analysis of variance (ANOVA) Results: The mean granulocyte count significantly (P , .01) decreased from 10,017 cells/mL (data range 3,001–20,170 cells/ mL) before the first radiation treatment to 3,250 cells/mL (820–4,400 cells/mL) at week 5 (P , .01). Three weeks after this nadir, the mean increased to 10,150 cells/mL (900–26,700 cells/mL). The hematocrit did not change (36–38%). Thrombocytopenia (,100,000/mL) occurred in 10 dogs. Two dogs died because of complications associated with thrombocytopenia. No significant difference in toxicity was found between the 6 Gy and 8 Gy group. Conclusions and Clinical Importance: HBI was myelosuppressive but effects were short term and resolved in 22 of 24 dogs. Further studies are needed to elucidate the safety and role of HBI in the treatment of dogs with lymphoma. Key words: Cancer treatment; Dogs; Myelosuppression; Radiation therapy.
hemotherapy is considered the mainstay of treatment for canine lymphoma. Multi-agent drug protocols induce remission in 80–90% of dogs within 1 to 3 weeks, and median remission times of 9 to 12 months are commonly reported.1 Relapses are frequent, and death caused by disease arising from the progression of drug-resistant disease is expected. Half-body irradiation (HBI) has been applied in an attempt to improve remission and survival times in dogs.2,3 The goal of HBI is to irradiate a large area of the body in a single exposure, while limiting bone marrow toxicosis. The hypothesis in human beings is that unirradiated bone marrow will circulate and repopulate the areas that are lethally irradiated if there is an appropriate interval between the 2 half-body doses.4 The use of HBI in people began as a means of disease palliation but has also been reported as treatment for disseminated cancers.5 Adverse effects in people include acute radiation sickness with nausea, vomiting, and diarrhea; and myelosuppression.6 HBI has been studied in normal dogs,7 in dogs with lymphoma as the primary treatment modality,8 and in dogs with lymphoma as a consolidation treatment in an attempt to improve remission rates and survival times.2,3
C
From Gulf Coast Veterinary Specialists, 1111 West Loop South, Suite 150, Houston, TX 77027. This study was presented in part in abstract form at the Veterinary Cancer Society 25th Annual Meeting, Huntington Beach, CA, October 2005. Reprint requests: Kevin Hahn, DVM, PhD, Gulf Coast Veterinary Specialists, 1111 West Loop South, Suite 150, Houston, TX 77027; e-mail:
[email protected]. Submitted April 11, 2006; Revised May 25, 2006; Accepted July 14, 2006. Copyright E 2006 by the American College of Veterinary Internal Medicine 0891-6640/06/2006-0017/$3.00/0
Adverse effects described are tumor lysis syndrome, myelosuppression, and radiation pneumonitis.8 Myelosuppression was mild and infrequent in dogs treated with induction chemotherapy before HBI. Thirteen percent of dogs developed a fever after cranial HBI and had an unremarkable CBC, other than thrombocytopenia, and complications associated with neutropenia or thrombocytopenia were not reported.2 In a study involving 6 dogs with HBI interposed in a cyclophosphamide, doxorubicin, vincristine, and prednisonebased protocol, neutropenia occurred in half of the dogs after radiation therapy, with no reported complications and or treatment-related thrombocytopenia.3 Despite these reports describing technique and efficacy of HBI in dogs with lymphoma, few give guidance to the distinctive toxicoses and associated complications of HBI in combination with chemotherapy. Our objective was to describe the hematologic effects of HBI as a consolidation or rescue therapy on a series of dogs with lymphoma during and after an HBI protocol. Based on normal bone marrow kinetics, the experience of others, and the observations made in an earlier subset of dogs, we hypothesized that HBI would appreciably but transiently affect the myelocytic and erythroid variables as observed by serial analysis of CBCs.
Materials and Methods The medical records of 37 dogs with multicentric lymphoma treated with both HBI and chemotherapy during the years of 2002 and 2003 were reviewed. Dogs that did not receive radiation therapy to both halves of the body or dogs in which hematologic data were not available were excluded from the study. A total of 29 records were suitable for review. Most dogs had diagnostic testing that included physical examination, survey radiographs of the chest, radiographs with or without ultrasound examination of the abdomen, a CBC, serum
Hematologic Changes with Irradiation chemistry analysis, and urinalysis. Bone marrow aspiration was not performed in any dog in this study. Dogs had to have no abnormalities on CBC to be eligible for the first fraction of half-body radiotherapy. Blood smears were performed in some dogs at various times, and not all results were available for analysis. HBI was used as consolidation therapy after induction of remission by a multi-agent chemotherapeutic protocol or as rescue therapy in addition to chemotherapy. All dogs were anesthetized with propofol and isoflurane during delivery of HBI. A 4 MV linear accelerator was used to deliver irradiation. The xyphoid process was marked and a line drawn transversely and dorsally was used as the defining mark between cranial and caudal body halves. Dogs were placed in lateral recumbency, and parallel opposed fields were used. A fall off of 4 cm was established both dorsally and ventrally. No organs were blocked out on any field in any dog and the limbs were manipulated, so that peripheral nodes were included in the radiation field. The source-to-surface distance varied, as large dogs needed extended fields. The half of the body treated first was determined based on tumor burden, with the half of the body with increased burden being treated first. Two different HBI protocols were used. In the first protocol, 4 Gy was delivered twice, 24 hours apart, to one half of the body and repeated on the other half of the body 1 month later for a total of 8 Gy to each half of the body. In the second protocol, 6 Gy was administered once in a single fraction to one half of the body and repeated on the other half of the body 1 month later. HBI was administered at week 0 (first body half) and week 4 (second body half). Complete remission (CR) was defined as no physical or clinical evidence of disease before irradiation but after induction with chemotherapy, before the second HBI, and after the second HBI. Partial remission (PR) was defined as a .50% reduction in lymph node size but ,100% reduction of measurable disease; stable disease (SD) was defined as a ,50% reduction in lymph node size or no change; and progressive disease (PD) was defined as increasing lymph node size or progressive clinical disease. Toxicoses were graded based upon the Veterinary Cancer Oncology Group Common Terminology Criteria for Adverse Events.9 The hematologic values obtained from a CBC were analyzed between the various sample collection periods (before each halfbody fraction, after the last half-body fraction) by repeated measures ANOVA.a A P value of #.05 was considered significant.
1399
Table 1. Number of dogs administered each chemotherapeutic agent before the first HBI fraction and days between administration of drug and HBI. $14 days 10–13 days 7–9 days
,7 days
Chemotherapeutics
n
n
n
n
Doxorubicin Cyclophosphamide CCNU Vincristine Asparaginase Mitoxantrone
9 1 1 0 2 2
1 1 0 0 1 0
0 2 0 1 4 0
0 0 0 0 4 0
CCNU, 1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea.
week 5; 15, at week 6; 7, at week 7; and 9, at week 8. Various schedules of chemotherapy were used to induce remission in dogs with lymphoma before the use of HBI. The chemotherapeutics used were asparaginase (n 5 26 dogs received at least 1 dose), vincristine (n 5 17), cyclophosphamide (n 5 17), doxorubicin (n 5 11), lomustine (n 5 1), mitoxantrone (n 5 2), and chlorambucil (n 5 1). All 29 dogs had chemotherapy with the intent to induce remission before the first HBI (Table 1). HBI was used as consolidation in 23 dogs and as rescue therapy in 6 dogs. The 8 Gy group included 15 dogs and the 6 Gy group, 14 dogs. The values reviewed from the CBCs obtained before the first HBI treatment (week 0) were above the lower limit of normal. Granulocyte count in blood decreased significantly (P , .01) 1 week after the second radiation treatment and returned to baseline by week 8 (Fig 1). Six of the 29 dogs (20%) had a granulocyte count of ,3,000/mL one or more times after the first HBI (Table 2). The hematocrit remained unchanged throughout the study period. Eight of 29 dogs in the study, 28%, had thrombocytopenia (Table 2). A substantial decrease in thrombocyte count
Results Dogs were 5.5 to 12.6 years of age at the time of initial examination. There were 14 spayed female dogs, 13 castrated male dogs, and 2 intact male dogs treated. Dog breeds included 14 mixed breeds, 3 Boxers, 2 Cocker Spaniels, 2 Scottish Terriers, and 1 each of the following breeds: Irish Setters, Basset Hounds, Vizslas, Rottweilers, Catahoulas, Cockapoos, Golden Retrievers, and Labrador Retrievers. HBI was not used as the sole treatment modality in any dog. All dogs received chemotherapy before radiation treatment. Chemotherapeutic drugs used to induce or maintain remission included vincristine,b cyclophosphamide,c doxorubicin,d asparaginase,e mitoxantrone,f lomustine,g and chlorambucil.h The chemotherapeutic drugs administered between radiotherapy half-body fractions were limited to asparaginase (n 5 21) or vincristine (n 5 2). All 29 dogs had a CBC at week 0, just before treatment. Eleven dogs had a CBC at week 1; 24, at week 2; 5, at week 3; 22, at week 4; 7, at
Fig 1. Granulocyte count (mean) in peripheral blood of dogs 5 radiation treatment. *P , .05 compared treated with HBI. with week 0.
O
1400
Axiak et al
Table 2.
Toxicosis encountered with grading according to the VCOG-CTCAE.
Grade 1 Toxicity Type
n
Granulocytopenia Thrombocytopenia
Anorexia
Diarrhea
Vomiting
Grade 2
Grade 3
n
Grade 4
n
n
3 (1,500–3,000/mL) 9 (100,000–150,000 / mL) 7 (coaxing or dietary change needed) 0 (.2 stools per day over baseline)
1 (1,000–1,499/mL) 2 (50,000–99,000/mL)
2 (500–999/mL) 6 (25,000–49,000/mL)
0 (,500/mL) 7 (,25,000/mL)
7 (oral intake altered ,3 days, no significant weight loss) 4 (2–6 stools over baseline, IV or SC fluids ,24 hours)
0 (life threatening, .5 days duration) 0 (life threatening, hemodynamic collapse)
0 (,3 episodes in 24 hours)
2 (3–5 episodes in 24 hours, fluids for ,24 hours)
0 (3–5 day duration, tube feeding) 1 (.6 stools, incontinence, IV fluids and hospitalization .24 hours) 6 (.5 episodes per day or .4 days duration)
0 (life threatening, hemodynamic collapse)
Grade 5 n N/A N/A
0 (death)
0 (death)
0 (death)
VCOG-CTCAE, Veterinary Cooperative Oncology Group–common terminology criteria for adverse events; N/A, not applicable.
(defined as ,100,000 platelets/mL) was not observed until week 2 (n 5 1 dog), 3 (n 5 1 dog), 5 (n 5 3 dogs), 6 (n 5 2 dogs), and 7 (n 5 1 dog) after the first half-body fraction (P , .01). The nadir of the mean thrombocyte count was 105,000 platelets/mL at week 5 (Fig 2). Two dogs died due to severe, unresolving thrombocytopenia. There were no significant differences detected between those dogs receiving 8 Gy and those receiving 6 Gy. After chemotherapy, and before the first HBI, 24 dogs were in CR, 4 in PR, and 1 had SD. Before the second HBI, 22 dogs were in CR, 1 dog had SD, and 6 dogs had PD. After the second HBI, 26 dogs were in CR, 1 dog had SD, and 2 dogs had PD. The duration of remission, defined as first remission after induction chemotherapy to first relapse, ranged
from 6 to 614 days, with a median of 173 days. Survival times (defined as time of diagnosis to time of death or loss of follow-up) ranged from 30 to 1,218 days, with a median of 300 days. Other adverse effects of HBI recorded were mild and included vomiting and diarrhea. The signs lasted no longer than 2 to 3 days and all but 1 of the dogs responded to outpatient supportive care (see Table 2). No severe acute gastrointestinal adverse effects developed in any dog on either the 6 Gy or 8 Gy protocol in the 9-week observation period. Restlessness and decreased appetite occurred in 1 dog and resolved with outpatient care. Neither tumor lysis syndrome nor acute radiation pneumonitis occurred in any dog in this study. Due to the variability in chemotherapeutic protocols, we were unable to reliably calculate statistical significance of the type, dose, or protocol of chemotherapy used.
Discussion
Fig 2. Thrombocyte count (mean) in peripheral blood of dogs treated with HBI. 5 radiation treatment. *P , .05 compared with week 0.
O
In normal dogs, the nadir for decreased granulocyte count occurs 7 days after the second HBI (week 5),7 and in dogs with lymphoma and HBI as the sole treatment, the granulocyte count nadir occurs at week 2.8 Both of these studies used a radiation dosage (7–8 Gy) comparable to the present study. The nadir for the decreased granulocyte count occurred 7 days after the second halfbody radiotherapy treatment (week 5) with either a 6 Gy (single fraction) or 8 Gy (2 fractions) protocol in the current study. While this nadir is statistically significant (P 5 .01), there was no clinical significance, as none of the dogs had clinical signs of sepsis. This finding suggests that bone marrow kinetics differ between dogs with lymphoma and dogs with lymphoma treated with chemotherapy. A severe, marked, and delayed onset thrombocytopenia was observed in this study. Thrombocytopenia did not occur until week 3 of the observation period in 1
Hematologic Changes with Irradiation
dog, and weeks 5 through 8 in 10 dogs. This is in contrast to that described in normal dogs, which developed a thrombocytopenia nadir 10 to 12 days after the first HBI. In that study, platelet counts returned to normal 25 days after the first HBI, and 32 days after the second treatment.7 In dogs with lymphoma with HBI as the sole treatment, thrombocytopenia is described as transient and the nadir is seen 10 to 14 days after the first treatment.8 Although we did not have sufficient data after week 8 to evaluate platelet recovery in all dogs, it is important to note that 2 dogs died of complications associated with thrombocytopenia. Both of these dogs had HBI (8 Gy group) with the intent of rescue therapy and had been intensively treated with chemotherapy beforehand. Both dogs had normal CBCs before HBI. Causes of their severe and nonrecovering thrombocytopenia might be stem cell damage from chemotherapy before HBI, HBI itself, or resistant lymphoma infiltration into the bone marrow. Mean hematocrit for all dogs was normal (33–55%) before the first HBI, and the mean did not significantly change throughout the 9-week study period. As the erythrocyte lifetime is 120 days, potential reasons for this stable value include the end point of the study at 9 weeks and that erythroid stem cells were not negatively affected by combination chemotherapy and HBI. The only chemotherapeutics used during radiation therapy were asparaginase and vincristine. We theorize that because asparaginase has a bone marrow sparing quality, it did not influence our interpretation of the hematologic data. Vincristine tends to be less myelosuppressive than many other chemotherapeutics, and only 2 dogs received vincristine between HBI treatments. No detectable difference in hematologic toxicity was found between the 6 Gy and 8 Gy groups. The median survival time of dogs in the current study was lower than in previous studies. However, this variable is difficult to interpret based on the variability of chemotherapeutic protocols and the fact that 6 dogs in this study underwent HBI for the purpose of rescue rather than consolidation. Also, 6 dogs lost to follow-up were censored (1 dog is still alive at a follow-up time of 1,218 days), and 19 dogs had chemotherapy after HBI. Although our study is limited by variation in chemotherapy protocols and the lack of bone marrow samples before irradiation, all dogs in this study had normal CBCs before radiation treatment, suggesting adequate bone marrow function. Dogs in this study that had HBI for the purpose of consolidation had a low incidence of adverse effects, with mild gastrointestinal upset being the most common noted. Two dogs in this study that had had HBI for the purpose of rescue died due to complications associated with thrombocytopenia after the second fraction, and caution needs to be exercised when using HBI for this purpose.
1401
The results of our hematologic assessment of HBI imply that although the bone marrow is affected by treatment, it can recover in most dogs. In contrast to other studies, no part of the body was blocked in either the cranial or caudal HBI fraction. Thus, our protocol as described in this study may be regarded as safe when used for consolidation in the treatment of lymphoma. However, caution should be exercised for the purpose of rescue and a bone marrow aspirate before HBI may be warranted. We conclude that HBI is myelosuppressive, and there is sufficient data to warrant further studies elucidating the therapeutic role of HBI in dogs with lymphoma in the concurrent or adjuvant setting.
Footnotes a
GraphPad Prism, Version 4.0, GraphPad Software, San Diego, CA b Vincristine, Gensia Sicor Pharmaceuticals Inc, Irvine, CA c Cyclophosphamide, Bristol-Myers Squibb Co, Princeton, NJ d Doxorubicin, Ben Venue Laboratories Inc, Bedford, OH e Asparaginase, Merck & Co, West Point, PA f Mitoxantrone, Immunex, Seattle, WA g Lomustine, Bristol-Myers Squibb Co, Princeton, NJ h Chlorambucil, GlaxoSmithKline, Research Triangle Park, NC
References 1. Vail DM, MacEwen EG, Young KM. Canine lymphoma and lymphoid leukemias. In: Withrow SJ, MacEwen EG, eds. Small Animal Clinical Oncology, 3rd ed. Philadelphia, PA: WB Saunders; 2001:558–590. 2. Williams LE, Johnson JL, Hauck ML, et al. Chemotherapy followed by half-body radiation therapy for canine lymphoma. J Vet Intern Med 2004;18:703–709. 3. Gustafson NR, Lana SE, Mayer MN, et al. A preliminary assessment of whole-body radiotherapy interposed within a chemotherapy protocol for canine lymphoma. Vet Comp Oncol 2004;2:125–131. 4. Rees GJ. Mechanism of total body irradiation in lymphoma. A hypothesis. Acta Radiol Oncol 1982;21:421–425. 5. Fitzpatrick PJ, Rider WD. Half-body radiotherapy. Int J Radiat Oncol Biol Phys 1976;1:197–207. 6. Poussin-Rosillo H, Salazar SM, Amin PA, et al. Palliative half-body irradiation, single and fractionated doses. Am J Clin Oncol (CCT) 1985;8:172–177. 7. Laing EL, Fitzpatrick PJ, Mosseri AM, et al. Half-body radiotherapy, evaluation of technique in normal dogs. J Vet Intern Med 1989;3:96–101. 8. Laing EL, Fitzpatrick PJ, Binnington AG, et al. Half-body radiotherapy in the treatment of canine lymphoma. J Vet Intern Med 1989;3:102–108. 9. Veterinary Co-operative Oncology Group, Common terminology criteria for adverse events (VCOG-CTCAE) after chemotherapy or biologic antineoplastic therapy in dogs and cats. Vet Comp Oncol 2004;2:194–213.
