Rescue Chemotherapy for Canine Lymphoma: What Next?
ACVIM 2008
Jeffrey N. Bryan, DVM, MS, PhD, DACVIM (Oncology)
Pullman, WA, USA


When primary chemotherapy protocols will no longer maintain a dog in remission from lymphoma, rescue protocols must be designed. It can be difficult to predict which drugs will be effective and confusing to implement in clinical practice. There are few algorithms published that will effectively rescue even a small majority of patients. The University of Wisconsin, Madison 25-week protocol (UW-25), a CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) protocol, reported a 100% success rate at effecting second remission in patients who had completed the entire protocol in remission the first time.1 This protocol is the author's preferred protocol, as its success at first rescue simplifies the initial decision of chemotherapy choice. Beyond this, however, selection of chemotherapy must be guided by the drugs already administered to the dog; the dog's ability to tolerate each class of chemotherapy agent; the client's willingness to risk toxicity, incur cost, and risk failure; and the practice's familiarity with the available agents. No one agent has consistently outperformed others in the literature. It is helpful to develop a working relationship with an oncologist in your area to become accustomed to their preferred rescue agents so that patients can transfer seamlessly back and forth between the general and referral practices. This lecture will discuss most of the agents currently available for lymphoma rescue, their applications, potential pitfalls, and the literature that supports their use.

Successful rescue depends, in part, on early identification of loss of remission. Instructing owners in lymph node palpation and regularly rechecking patients at your practice will often allow detection of loss of remission in its early stages. The greater the disease burden, the higher the risk of therapy-related complications. In the author's experience, patients presenting with relapse who are ill (substage b) are less likely to re-enter remission. Staging is performed in the rescue setting only so far as necessary for safe therapy. Drug selection must be based on several factors. First, toxicity and patient safety are paramount. A complete blood count should be performed prior to administration of any myelotoxizc drug, every time. Appropriate serum chemistry monitoring should be performed. Second, always think two steps down the road in drug choice. If this drug does not work, how soon will another drug be safe to administer? Third, discuss dosing frequency, cost, potential toxicity, and expectation of success with clients. A rule of thumb is that subsequent remissions will be half as likely to occur and last half as long. There reaches a point in every lymphoma case where there are diminishing returns on investment of resources, time, and potential toxicity. Engage your client in a discussion of quality of life. Help define end-points beyond which the dog's life will not be prolonged to ensure that suffering is bearable for the patient. Finally, always consider the dog's entire well-being. Address the need for nutrition, mobility, infection control, and pain management.

Single-Agent Rescue

The author typically reserves L-asparaginase for lymphoma rescue. This drug is generally very well tolerated and usually effective if it has not been administered multiple times previously. The primary toxicity associated with L-asparaginase is anaphylactic reaction, as it is a protein. This is more likely to occur in dogs who have received the drug in the past. For repeat administration, pre-medicate the dog with diphenhydramine and possibly dexamethasone. There is also some risk of pancreatitis with each administration of this drug, so a high index of suspicion should be maintained for this condition if vomiting or diarrhea are noted after administration. The risk of pancreatitis appears to be very low. The myelotoxicity of L-asparaginase is extremely low, making it highly suited to induction of patients out of remission. Almost any other drug can be given at almost any time after L-asparaginase without concern for additive toxicity. There is some concern that L-asparaginase may prolong the metabolism of some drugs, such as vincristine, in the first few days after its administration.2 This does not appear to be a clinical problem in most cases. The cost of L-asparaginase is moderately high. It may be dosed as often as weekly, but is effective as a single agent for only a limited period. The author uses a single dose for re-induction, and then administers a different drug to maintain remission.

Vinblastine is a vinca-alkaloid class of drug that may be effective as a single agent, even when vincristine has failed. It is mildly more myelotoxic than is vincristine. The dosage in dogs is 2mg/m2, which is significantly different from vincristine. The two must never be confused, as this dose of vincristine could easily be lethal. Vomiting and diarrhea is rare with vinblastine. The dosing schedule is once weekly for this drug, so, in the event it is not effective, a different drug can be chosen the following week. The cost of vinblastine is low. Often, it will be combined with prednisone and L-asparaginase in the author's practice. Vinblastine is an MDR-1 substrate with increased risk of side-effects in dogs with mutant MDR-1 such as collies and collie-like dogs.3

Lomustine (CCNU) is an alkylating drug that may also be effective as a single agent. In one study of 43 pre-treated dogs with lymphoma, this drug induced complete remission (CR) in 27% of dogs, partial remission (PR) in 7% with a 3 month average remission duration.4 The remission was, on average, 4 months in duration with dogs in CR.4 This was not improved by the addition of prednisone in another study with a median response duration of only approximately 1.5 months.5 A recent paper combined CCNU with L-asparaginase and prednisone, resulting in a 52% CR and 35% PR rate.6 The CR was durable, at an average of 4 months, with an overall duration of remission of 2 months.6 The toxicity of CCNU is primarily myelotoxicity. Rarely, vomiting or diarrhea will be observed. The drug can cause profound neutropenia and thrombocytopenia that can last for up to three weeks. Thrombocytopenia that does not rebound in three weeks is cause for discontinuing the drug. Some patients may benefit from prophylactic trimethoprim-sulfa administration to avoid sepsis during neutropenic periods. The author's experience is that sepsis is a rare event. The drug can be hepatotoxic, so ALT should be monitored prior to, and at each administration of therapy.7 The dosage in dogs recommended by the author is 60mg/m2 PO every three weeks. When administering CCNU, one must assume that no other cytotoxic drug can be administered for up to three weeks. Rarely, a patient could be treated at two weeks. This becomes a factor in selecting this drug, as failure results in delay of further rescue. The cost of the drug is moderate. The drug can be scripted to pharmacies. It is supplied in 10mg and 40mg capsules. The calculated dose should be rounded down to the nearest combination of capsule sizes. This is a very common rescue agent for lymphoma.

Doxorubicin was examined as a rescue agent in 12 dogs who had failed chemotherapy, some already containing doxorubicin. Of these dogs, 42% experienced a CR and 33% a PR with an average duration of 5 months.8 These results suggest that this is an excellent rescue agent from the standpoint of patient response. No dog who failed to respond to the first dose then responded to a second.8 Another paper examined the use of doxorubicin and dacarbazine together. Of these 15 dogs, 53% experienced a CR and 33% a PR, with a lesser duration of only approximately one month.9 Toxicity of doxorubicin is primarily myelosuppression, vomiting, and diarrhea. Anti-emetic medication is necessary in approximately 1/3 of patients, in the author's experience. The addition of dacarbazine did not appear to worsen the toxicity associated with doxorubicin administration. Doxorubicin is also cardiotoxic, particularly when administered to a cumulative dose > 180mg/m2. For this reason, its use may be limited by total cumulative dose. There is a balance to consider, however, in extending life by treating lymphoma without limiting life by causing cardiotoxicity. Doxorubicin should be administered at a dosage of 30mg/m2 (1mg/kg in dogs < 15kg), diluted in at least 30mL saline, over at least 30 minutes, every 3 weeks. After administration of doxorubicin, no other myelotoxic therapy is possible for two to three weeks. Clients must understand the cardiotoxic potential of this drug, and cardiac monitoring in the form of periodic echocardiograms should be performed. If one dose of the drug does not induce remission, a different drug should be considered. The cost of doxorubicin is moderate. The technical handling of this drug is somewhat complex, as it must be delivered into a well-placed catheter. Extravasation will cause massive necrosis and extreme damage to the limb. Great care must be taken in its administration.

Mitoxantrone is synthetic anthracycline, similar to doxorubicin. It has been examined in two papers, looking at a total of 49 dogs. The CR rate ranged from 38% to 47% and the PR rate from 26% to 47% yielding an average duration of remission of three to four months.10,11 This response rate is lower than that of doxorubicin. Mitoxantrone is more myelotoxic than doxorubicin, but causes less gastrointestinal toxicity. Mitoxantrone is administered at a dosage of 5mg/m2, diluted in 30mL saline, over 20 minutes. It is less of a vesicant than doxorubicin, but care should be taken to ensure intravenous administration. The cost of mitoxantrone has decreased recently, but remains high relative to doxorubicin. This drug is generally well tolerated, and is a good choice for rescue if cost is not a determining factor.

The anthracycline actinomycin-D has been used as a single-agent rescue drug. The original paper described a 33% CR and 44% PR rate with a median duration of 1.5 months in nine dogs.12 A subsequent paper found no response, whatsoever, in 25 dogs.13 It is difficult to account for the vastly different outcomes in these two papers. A recent abstract, presented at the 2006 Veterinary Cancer Society Annual Meeting, found a 40% CR and 1% PR rate in 65 dogs treated with actinomycin-D as a rescue agent (Bannink E, personal communication). These results suggest that this drug is appropriate for use as a rescue agent for canine lymphoma. The author uses a dosage of 0.7 mg/m2, diluted in 20mL saline, and administered slowly intravenously. This drug is generally well tolerated, with rare gastrointestinal toxicity seen. Myelotoxicity can occur. The cost of this drug has recently become very high. This has limited its use in the author's practice. The cost may decrease in the future, so the actinomycin-D should be considered among the rescue agents.

The drug etoposide (VP-17) was administered to 13 dogs with lymphoma, 12 treated with prior chemotherapy.14 Of these, only two had a response, one a CR.14 Eleven of the dogs experienced an acute, pruritic cutaneous reaction. This was judged to be a response to the carrier agent in the drug, polysorbate 80. Given this reaction and the poor response rate, the authors concluded that this drug has little effectiveness against lymphoma. It does not appear to be a good rescue drug.

Multi-Drug Rescue

Of multi-drug protocols used for rescue, MOPP (mechlorethamine, vincristine, procarbazine, prednisone) has one of the best track records. In a study of 117 dogs, a CR rate of 65% and a PR rate of 31% were observed.15 The median duration of remission for dogs with a CR was 2 months.15 Most interesting was that nine of 76 dogs had longer rescue remissions than first remission.15 This finding is unique among papers describing rescue chemotherapy. It has resulted in several other multi-drug protocols attempting to achieve the same result with drugs other than mechlorethamine, which has been periodically difficult to obtain. Replacing mechlorethamine with CCNU (LOPP) resulted in a 27% CR and 25% PR rate in 40 dogs. The CR duration was a median of 4 months. The protocol was significantly toxic, with neutropenia, thrombocytopenia, and gastrointestinal toxicity predominating.16 The same paper examined BCNU in place of mechlorethamine (BOPP), achieving similar results of 29% CR and 21% PR with a median CR duration of 4 months.16 MOPP itself can result in significant gastrointestinal toxicity and myelotoxicity. Mechlorethamine is readily obtained, but extremely expensive. Procarbazine can also be very expensive, but when compounded by a licensed pharmacy is generally affordable. Timing of chemotherapy administration is defined by the protocol in this case, so there is less concern for alternatives in the event of failure. Generally, another drug could be given within one to two weeks of administration of these drugs. An abstract presented at the 2005 Veterinary Cancer Society Annual Meeting suggested that this protocol may be useful for dogs with T-cell lymphomas (Brodsky E, personal communication). For this reason, this protocol may be most appropriate for rescuing T-cell lymphomas treated initially with CHOP protocols. MOPP, LOPP, and BOPP may be quite toxic for some dogs. Clients should be educated carefully on the cost and risks of these protocols. From a tumor biology standpoint, combination therapy is most desirable, so MOPP may be one of the best choices among rescue options.

Combination of temozolamide or dacarbazine and an anthracycline, either doxorubicin or actinomycin-d, were administered to 53 dogs with relapsed lymphoma.17 Objective responses were seen for 72% (median duration 40 days) and 71% (median duration 50 days) of dogs for temozolamide and dacarbazine, respectively.17 Temozolamide was reported to be less myelotoxic and more convenient to administer than dacarbazine. The selection of anthracycline made no significant difference in outcome.

Radiation Rescue

Radiation therapy may play a palliative role in rescue therapy. It is most effective at alleviating discomfort cause by grossly enlarged regional nodes. Half-body radiation as a rescue therapy has not proven effective. Rescue chemotherapy can be interspersed around radiation, but the combination may be particularly myelotoxic. Also, drugs such as doxorubicin may result in dramatically greater radiation effects on normal tissue if administered concurrently. Close coordination with a specialist in your area will facilitate transfer between your practice and a radiation facility.


There is no single, best drug to begin rescue of lymphoma. Often, the choice is made based on minimizing toxicity and cost, and ease of administration after the effort and expense of the initial protocol. If these are not factors, begin with L-asparaginase. This can be followed by vinblastine, the therapeutic spectrum of which does not overlap entirely with vincristine. If this does not work, doxorubicin may be administered if the patient has received four or fewer prior doses. After six doses of this drug, the likelihood of cardiotoxicity becomes greater. Clients should be educated carefully about this risk. A drug that is effective should generally be continued as long as the patient is in remission and toxicity has not manifested. If a client is willing to accept greater expense and risk more severe toxicity, MOPP may offer the best chance of a durable remission. Because two of the drugs in the protocol are alkylating agents novel to a dog that was treated with CHOP previously, multi-drug resistance may be overcome. The alternative LOPP and BOPP protocols have not proven as effective as the first report of MOPP. Some dogs will have less aggressive forms of lymphoma and enter durable remissions for a second time. If a dog remains in remission after two or three doses of doxorubicin, it should be temporarily discontinued until remission is lost. If a dog remains in remission using MOPP or similar protocol, discontinue treatment again after six months and observe for relapse. Generally, however, dogs will not respond this well, and remission will be lost while being treated with the rescue agents. Select the next agent based on alternating classes of drugs and client willingness to tolerate toxicity.

Chemosensitivity assays have not been shown to be beneficial in selecting rescue drugs. Much like bacterial antibiotic sensitivity assays, chemosensitivity assays culture cells and test sensitivity to various chemotherapy drugs. In a prospective study of this assay in dogs with relapsed lymphoma, only 19% of dogs experienced a CR or PR after treatment with the suggested drug.18 This is no better than selecting a rescue agent at random, so is not worth the effort and expense. Until a better test comes along, rescue chemotherapy will be based on matching drugs with the needs of the patient and client.


Rescue of cats with relapsed lymphoma has received very little attention in the literature. Generally, the clinical course of feline lymphoblastic lymphoma is more aggressive than canine. Once relapse occurs after primary therapy, it is the author's experience that rescue is difficult to achieve. If the cat has completed its initial protocol and is not receiving chemotherapy at the time of relapse, these drugs should be attempted on the previous schedule. Doxorubicin can be administered as a single agent if the cat has not received this previously. Unfortunately, even in newly diagnosed lymphoma, doxorubicin results in relatively few CR.19 Radiation therapy may be used to palliate regional clinical signs, similar to dogs. Be sure to check dosages and toxicity with your oncologist before administering drugs with which you are not familiar.


1.  Garrett LD, et al. J Vet Intern Med 2002;16(6):704.

2.  Northrup NC, et al. J Vet Intern Med 2002;16(5):570.

3.  Mealey KL, et al. J Am Vet Med Assoc 2003;223(10):1453.

4.  Moore AS, et al. J Vet Intern Med 1999;13(5):395.

5.  Sauerbrey ML, et al. J Am Vet Med Assoc 2007;230(12):1866.

6.  Saba CF, et al. J Vet Intern Med 2007;21(1):127.

7.  Kristal O, et al. J Vet Intern Med 2004;18(1):75.

8.  Calvert CA, et al. J Am Vet Med Assoc 1981;179(10):1011.

9.  VanVechten M, et al. J Vet Intern Med 1990;4(4):187.

10. Lucroy MD, et al. J Vet Intern Med 1998;12(5):325.

11. Moore AS, et al. J Am Vet Med Assoc 1994;204(12):1903.

12. Hammer AS, et al. J Vet Intern Med 1994;8(3):236.

13. Moore AS, et al. J Vet Intern Med 1994;8(5):343.

14. Hohenhaus AE, et al. J Vet Intern Med 1990;4(5):239.

15. Rassnick KM, et al. J Vet Intern Med 2002;16(5):576.

16. LeBlanc AK, et al. Veterinary and Comparative Oncology 2006;4(1):21.

17. Dervisis NG, et al. J Am Vet Med Assoc 2007;231(4):563.

18. Henry CJ, et al. J Am Anim Hosp Assoc 2001;37(2):165.

19. Kristal O, et al. J Vet Intern Med 2001;15(2):125.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Jeffrey Bryan, DVM, MS PhD, DACVIM (Oncology)
Pullman, WA