Naproxen Test for Neoplastic Fever May Reduce Suffering

Introduction

Although neoplastic fever (NF) is one of the major causes of fever in cancer patients, ¹ it is frequently under-diagnosed. ² Malignancy is the cause of fever in 17% of patients with fever of initially unknown origin. ³ In a prospective study of 477 febrile episodes among cancer patients, cancer itself was the cause of 27% of non-infectious episodes. ⁴ Other causes of non-infectious fever include drugs, invasive procedures, thromboembolic disease and blood transfusion.^1,4

Fever of neoplastic origin has been described in hematological, as well as solid tumors. ¹ In some tumors, such as renal cell carcinoma, its incidence may be as high as 30%. ⁵ The pathophysiology of NF is not clear, but is likely mediated by cytokines.^1,2 Its diagnosis is based on the exclusion of other causes of fever. A positive naproxen test may be a useful adjunctive test. ²

Evidence shows that NF can be managed effectively with nonsteroidal anti-inflammatory drugs (NSAIDs) and that naproxen may be particularly effective in some cases. ⁶ In patients with advanced cancer and limited life expectancy, it is important to control symptoms effectively and avoid unnecessarily burdensome treatments as much as possible. We describe a case of renal transitional cell carcinoma associated with NF that was controlled with a simple medication (naproxen) after a period of a relatively burdensome treatments and significant suffering.

Case Description

A 37-year-old female presented in December 2007 with three months of colicky left-flank pain. CT scan of the torso revealed a left renal mass with regional lymphadenopathy, a solitary hepatic lesion, and a pulmonary nodule. Tissue from a left nephrectomy showed poorly differentiated transitional cell carcinoma. Over the next eight months, her disease progressed despite two consecutive chemotherapy regimens. The decision was made to stop further cancer-modifying therapy and to focus on treating symptoms. In September 2008, she was hospitalized with fevers to 39°C, sweating, fatigue, and generalized arthralgias. Upon hospitalization, her Eastern Cooperative Oncology Group (ECOG) performance status was 3 (capable of only limited self-care, confined to bed or chair more than 50% of waking hours). ⁷ Clinical examination, laboratory investigations, and imaging studies failed to reveal a source of infection.

Two empiric antibiotic regimens were tried without success: meropenem (1.5 g/day, day 1 to day 4 of hospitalization) and piperacillin/tazobactam (13.5 g/day, day 5 to day 10). From day 1 to day 17, she received acetaminophen (2 g/day) with no noticeable impact on her fever or constitutional symptoms. There was temporary fever control following the initiation of indomethacin (100 mg/day, day 1 to day 5), betamethasone (14 g, once; day 7), and dexamethasone (16 mg/day, day12 to day 17). However, none of these medications led to significant improvement in her constitutional symptoms or performance status.

On day 17 of the patient's hospitalization, the palliative care team was consulted and started her on naproxen at 250 mg twice daily. The same day, the patient experienced a dramatic response, with complete lysis of fever and control of constitutional symptoms. The next day, she was discharged home with an ECOG performance status of 1 (restricted in physically strenuous activity but ambulatory and able to carry out work of a light or sedentary nature, e.g., light house work, office work). ⁷ One week later, she stopped the naproxen and had a recurrence of fever and other symptoms, which resolved again after restarting naproxen. The patient died of advanced disease within two months.

Discussion

Urological malignancies may be associated with a wide variety of paraneoplastic syndromes, including NF, which is commonly associated with renal cell carcinoma. ⁵ Few reports document NF in association with transitional cell carcinoma of the urinary bladder.^8,9 The current case may be the first case of isolated renal transitional cell carcinoma associated with NF to be reported in the English-language literature. In one report, a patient had synchronous renal cell carcinoma and transitional cell carcinoma of the renal pelvis. ¹⁰ However, NF in that case may have been the result of renal cell carcinoma. ⁵

For the diagnosis of NF, six-item criteria have been proposed: temperature >37.8°C at least once daily, fever duration >2 weeks, no evidence of infection (by physical examination, laboratory investigations, or imaging studies), no allergic mechanisms, no response of fever to empiric antibiotics for ≥7 days, and naproxen test resulting in complete lysis of fever, which is maintained while on naproxen. ² The naproxen test, first described in 1984, ¹¹ may be a sensitive diagnostic tool for NF with rare false-positive results. ² A suggested regimen for the test is to give naproxen 375 mg orally every 12 hours for a period of ≥36 hours. The test is positive for NF when there is complete and sustained lysis of fever while receiving naproxen. ²

Despite enthusiasm about the use of naproxen and other NSAIDs in NF, ² the results of NSAIDs in NF in general have been mixed. One randomized trial, which included 48 patients, compared indomethacin and diclophenac to naproxen in their efficacy in controlling NF. The investigators found the three NSAIDs equally effective in NF control, but that naproxen resulted in a more rapid effect. ⁶ The current case showed only temporary response to indomethacin, followed by relapse of fever. Minetto et al. reported the failure of metamizole to control NF in a case of pheochromocytoma. ¹²

Steroids, like NSAIDs, show mixed results. In our case, steroids resulted in a temporary reduction of fever, which then relapsed. In a retrospective study, Chang observed that steroids are less effective than naproxen in controlling NF. ¹³ In that study, treatment with naproxen resulted in complete lysis of fever in 36 (90%) out of 39 patients with NF. When 12 of these 39 patients received corticosteroids alone, only six (50%) had complete lysis of fever. ¹³

It is known that NSAIDs are not identical regarding their mechanism of action. ¹⁴ At the same time, the pathophysiology of NF should be more fully elucidated.^1,2 Further randomized controlled trials to compare the effect of different NSAIDs on NF could be useful. In addition, research to understand the pathophysiology of NF may lead to a more targeted approach.

While the current evidence supports the use of non-selective NSAIDs for NF control,^2,6 the benefits from these drugs should be weighed against their side effects. Unless contraindicated, selective COX-2 inhibitors may be used in patients at risk for gastrointestinal toxicity and those with thrombocytopenia. ¹⁵

Conclusions

In patients with advanced cancer and no further anti-cancer treatment options, it is acceptable to investigate reversible causes of infection, but it is also important to avoid unnecessary investigations and treatments that may cause suffering. In the case of the patient described in this article, it made sense to investigate potential sources of infection and to give empiric antibiotics for seven days. Given the data presented above, clinicians may want to initiate a trial of naproxen as the first-line NSAID for patients in whom no infectious cause of fever is identified and empiric antibiotics fail. In the current case, earlier use of naproxen might have decreased this patient's hospital stay by more than a week, reducing suffering and increasing her time at home during the last weeks of her life.