Clinico-Pathological Characteristics, Treatments and Outcomes of Patients with Dendritic Cell Sarcoma (DS)

Background: Dendritic cell sarcomas (DCS) mainly include follicular dendritic cell sarcoma (FDCS) and interdigitating cell sarcoma (IDCS). DCS are difficult to diagnose. There is no standard of care and research on DCS is sparse due to its rarity. We present our data on the clinico-pathological characteristics, treatments and survival of patients (pts) with DCS.

Methods: This is a retrospective analysis of 64 pts with confirmed diagnosis of DCS (FDCS n=58; IDCS n=6) evaluated at MDACC (1995 and 2015). Following IRB approval, pt data were collected on the clinico-pathological features, treatments and clinical outcomes. Histopath. and immunohistochemistry (IHC) data were collected.

Results: Median age at diagnosis was 51 yrs (range 16-77 yrs). 35 pts were female and 29 were male. 51% pts presented with localized disease and 49% with systemic involvement. Overall, 13 pts had nodal involvement, 21 had extra nodal disease and 21 had both. The abdomen was involved in 33/58 pts with liver (9/33) followed by spleen (5/33) as the most common extranodal sites of disease. 12 pts (19%) had pre-existing or subsequent autoimmune disease (3 had pemphigus, 3 thyroid disease and 2 had myasthenia gravis and 4 had other). 6 pts with FDCS had Castleman's disease (hyaline vascular). IHC markers in FDCS were clusterin, EGFR, vimentin, CD21, CD35 and CD23 in 90%, 81%, 87%, 81%, 80% and 64% samples. Ki-67 was > 25% in 44% pts. S-100 and CD68 were expressed in 11% and 28% respectively. For IDCS, markers included S-100 and CD68 in 100% and 83% pts. Electron microscopy was performed in 3 pts and revealed neoplastic cells with well-formed junctions and very prominent desmosomes. Molecular studies in 2 pts were negative for BRAF mutations. One pt had TP53 exon 8 and another had PTEN mutation.

50 pts had treatment information available, and 45 had response data. Initial treatments included surgery followed by adjuvant chemotherapy with/without radiation in (21/50), chemotherapy alone in (7/50), radiation with/without chemotherapy in 6 pts and surgery with radiation in 6 pts. The overall response rate (ORR) to initial therapy was 78% (35/45) - FDCS 78%; 32/41 and IDCS 3/4; 75%. Complete response (CR) was observed in 40% (18/45) - FDCS 41%; 17/41 and IDCS 1/4; 25%. Partial response rate (PR) was 38% (17/45) - FDCS 37%; 15/41 and IDCS 2/4; 50%. In pts who received chemotherapy, gemcitabine with taxanes (Gem-Tax) (n=12) and CHOP based regimens (n=12) were most commonly used and were associated with ORR of 83% (10/12) and 75% (9/12) respectively. The CR rate following chemotherapy (alone or following local therapy) with both Gem-Tax and CHOP was 42%. Among pts who received Gem-Tax with/without other modalities - 42% (5/12) were CR and 41% were PR (5/12) and 2 pts were non responders. For the 10 responders, 6 pts had surgery prior to chemotherapy, 3 had radiotherapy and 1 had chemotherapy alone. For pts who received CHOP based regimens with/without other modalities - 42% (5/12) were CR and 33% were PR (4/12) and 3 pts did not respond. For the 9 responders, 6 pts had surgery prior to chemotherapy, and 3 had radiotherapy. Eight pts were treated with miscellaneous regimen as first line, Ifos. with Adria. (3) and one each with R-COP, R-CHOP, lenalidomide with prednisone, imatinib and ICE with surgery or radiotherapy. Overall, 27 pts were alive (23 FDCS and 4 had IDCS) and 33 pts died (31 FDCS and 2 had IDCS) at the time of last follow up - (median 45 months; range 5-129 months). Overall, median progression free and overall survival was 33 and 50 months respectively. Furthermore, among the FDCS, survival outcomes were significantly shorter in pts with extra nodal disease as compared to nodal disease alone (Figure 1A-B- p=0.018 for PFS and p=0.030 for OS).

Conclusions: We present the largest single center report of pts with DCS. We observed marginally higher responses with gemcitabine/docetaxel combinations compared to CHOP based therapy, but limited numbers make comparisons difficult and prospective studies are needed to define optimal treatment in DCS. Combined modality treatment appears to result in improved outcomes. Moreover, we have shown that pts with nodal disease alone at the time of initial presentation have significantly longer survival. We are studying the genomic profile in DCS and identify potential targets for therapy.

Figure 1.

View largeDownload slide

Progression free and overall survival in patients FDCS

Figure 1.

View largeDownload slide

Progression free and overall survival in patients FDCS

Close modal