A Case of Capecitabine-Induced Discoid Lupus Erythematosus

Dear Editor:

Cutaneous lupus erythematosus (CLE) induced by oral capecitabine is rarely reported. We report a case of oral capecitabine-induced discoid lupus erythematosus (DLE). The use of regimens containing 5-fluorouracil (5-FU) before and after capecitabine did not cause any symptoms, but interestingly, DLE was induced after taking capecitabine, a pro-drug of 5-FU.

A 79-year-old female diagnosed with metastatic recto-sigmoid cancer took oral capecitabine after surgical resection and six cycles of chemotherapy using LF protocol (leucovorin plus 5-FU). Progressive skin lesions developed 5 months after starting capecitabine, presenting dusky erythematous indurated plaques with scales as malar rash features in the cheeks and discoid rashes on the forehead and philtrum. There was no skin lesion on the ears (Fig. 1). She didn't feel itching or irritating. The patient had no history of lupus erythematosus. A punch biopsy of the right cheek lesion revealed epidermal thinning and flattening with follicular plugging and basal vacuolar degeneration. There were lymphohistiocytic infiltrates around hair follicle and densely near the dermal-epidermal junction. Patchy inflammatory infiltrate extended into deep dermis. Mild interstitial mucin deposition with solar elastosis was identified (Fig. 2). Direct immunofluorescence stains revealed deposition of immunoglobulin M along the dermal-epidermal junction. Laboratory analysis presented positive for anti-Ro/SSA antibodies (>21.0 U/ml) and negative for antinuclear antibodies (ANA). There was no abnormality in the hematologic, biochemical blood tests. Capecitabine was discontinued after diagnosis of drug-induced DLE and the symptoms were improved within a month. Then, FOLFIRI protocol containing leucovorin, 5-FU, and irinotecan was used and no new skin symptoms appeared.

Fig. 1. (A~C) A 79-year-old female presented with dusky erythematous indurated plaques with scales as malar rash features. We received the patient's consent form about publishing all photographic materials.

Fig. 2. A punch biopsy specimen shows basal vacuolar degeneration with follicular plugging. In the dermis, there is dense lymphohistiocytic infiltration arranged along dermalepidermal junction and the infiltrate extends into the deep dermis. (A) Interstitial mucin deposition is present (H&E; ×40; inset, ×200). (B) A higher power view shows thinning and flattening of epidermis with hyperkeratosis (H&E; ×100).

Capecitabine is an oral fluoropyrimidine carbamate generating 5-FU preferentially in tumor tissue through exploitation of higher intratumoral concentrations of thymidine phosphorylase¹,². In comparison, when intravenous 5-FU is administered, the level of active 5-FU in the tumor is not greater than that in normal tissue². Thus, oral capecitabine has less systemic toxicity than intravenous 5-FU³.

There are a few reports of capecitabine-induced CLE, of which most of are subacute CLE. In particular, only one case showed DLE-like features⁴. In our case, though the rash was localized to the face with butterfly shape, prominent epidermal change, interstitial mucin and dense lymphocytic infiltration from dermal-epidermal junction to deep dermis suggested DLE. All of the reported cases of capecitabine-induced subacute CLE showed increased titer of anti-Ro/SSA antibodies, but the one case that showed discoid-like lesion was normal⁴. Drugs are capable of inducing CLE through Ro/SSA antibody dependent cell-mediated cytotoxicity and antibody production⁵. There are also reports that identified positivity for anti-histone antibodies in drug-induced subacute CLE, but the percent of positivity was lower than those in ANA or anti-Ro/SSA antibodies⁵.

In this case, discoid and malar rashes accompanied by an increased titer of anti-Ro/SSA antibodies after capecitabine treatment were improved shortly after capecitabine was stopped. Interestingly, although capecitabine is a precursor of 5-FU, no skin symptoms occurred when 5-FU was used before or after the onset of DLE induced by capecitabine. This case is meaningful for considering the unknown metabolic components acting on converting capecitabine to 5-FU of CLE induced by oral capecitabine.