Abstract
INTRODUCTION: The typical presentation of patients with low grade lymphoma consists of disseminated lymphadenopathy, absence of constitutional symptoms, normal LDH, low Ki-67 and PET scan with low SUVs (i.e. <14). We have identified a subset of atypical cases who present with one or more clinically aggressive features. We have named these cases clinically discordant indolent histologies (CDIH). The goal of this study is to identify these cases with CDIH in order to determine if their prognosis and clinical behavior are different from those with the typical low grade NHL presentation.
METHODS: We defined CDIH as any follicular grade 1-2, grade 3-A or small lymphocytic lymphoma (SLL) who meet at least one or more of the following conditions: constitutional symptoms, unexplained LDH elevation, PET SUV >14, Ki67 >30%, unusual areas of involvement for indolent NHL (bone, pleura, CNS, soft tissue, lung), necrotic areas seen in CT scan or discrete space occupying lesions in liver or spleen. We analyzed their failure free survival (FFS), overall survival (OS), rate of transformation to a high grade histology and correlation with FLIPI-2 prognostic score.
RESULTS: A total of 97 cases (86 follicular, 11 SLL) with a median follow up of 56 months were identified from our data base. Of these 97 cases, 46 met the criteria for CDIH. Figure 1 shows the FFS of cases with CDIH as contrasted with 51 without CDIH. As is evident from Figure 1, those with CDIH not only had a higher relapse rate but also showed a trend for earlier relapses (within 3 years), reminiscent of high grade NHLs. Figure 2 depicts the OS of CDIH cases which is significantly inferior. We also analyzed the risk of transformation. The rate of transformation of CDIH was 5/46 (11%) in contrast to 0/51 with non CDIH (p=.02). All episodes of transformation occurred early, between 6 to 35 months from diagnosis. In order to determine if there was an underlying lymphoma of high grade histology at diagnosis, 8 cases with CDIH underwent a second biopsy of a site suspected of harboring an aggressive NHL because of findings such as SUV of ≥14. None of the 8 cases showed evidence of a high grade NHL in the second biopsy. In nearly all (94/97=97%) treatment included rituximab in combination with chemo and 80% of these were given maintenance. Management consisted of: Treatment A (n=55)-non-doxorubicin regimens, mostly fludara or bendamustine based; Treatment B (n=19)-R-CHOP; Treatment C (n=23)-R-CHOP x 6 followed by fludara based regimen (FND) x 4. The latter was used primarily for CDIH. Of the 46 cases of CDIH, 31 were treated with a doxorubicin-rituximab combination (either Treatment B or C). Their OS at 8 years was 98 % vs 62% for those who received a non- doxorubicin regimen (Treatment A), p=0.014. Of 40 cases without CDIH who received Treatment A, only 1 has relapsed. FLIPI-2 prognostic score correlated poorly with CDIH: of 23 with high risk score, 12 had CDIH and of 7 with low risk, 3 were CDIH.
CONCLUSIONS: 1-Cases with CDIH have less favorable OS, FFS, and higher rate of transformation to a higher grade histology. 2-In spite of the fact that CDIH histologically is identical to an indolent NHL, it functionally behaves as an aggressive NHL with frequent early relapses. 3-FLIPI-2 prognostic score correlates poorly with CDIH. This is not surprising since FLIPI-2 score doesn't include B symptoms and LDH. 4-Confirmatory biopsies to rule out co-existing high grade NHL at diagnosis are not useful or recommended. 5-Those without CDIH do very well when treated with a non-doxorubicin regimen such as a fludarabine-rituximab containing combination while CDIH cases appear to fare better when treated with a doxorubicin-rituximab containing regimen. These conclusions have to be interpreted in light of the retrospective nature of the study.
Disclosures
No relevant conflicts of interest to declare.
FAS and Subsequent Therapies in Pancreatic Neuroendocrine Tumors