ICML 2017 | Chemotherapy-free approaches to treat Follicular Lymphoma – Meet the Professor Session

One of the first sessions that the LH attended at the 14^th International Conference on Malignant Lymphoma (ICML), taking place this week in Lugano, Switzerland, was a “Meet the Professor” session by J.W. Friedberg, MD, MMSc, from the Wilmot Cancer Institute, University of Rochester Medical Center, Rochester, NY, USA. The session focused on Follicular Lymphoma (FL) and how it can be treated with chemotherapy-free strategies.

The talk began with an overview of FL with Friedberg stating that FL is the most common type of indolent NHL; in the USA, more than 24,000 cases are diagnosed each year and this figure is rising. FL carries a variable clinical course and in some cases is incurable with conventional treatment resulting in numerous cycles of remission then relapse. Histological transformation can also occur; however, Friedberg eluded to other data that will be presented at the meeting suggesting that the incidence of histological transformation is decreasing. Moreover, it seems that median Overall Survival (OS) for patients is improving; when Friedberg began his career, median OS was around 6–8 years compared to today where median OS can exceed 20 years in patients with advanced stage disease.

However, there are a number of patients who are progressing and dying early, and so there is much more work to do. The speaker suggested that when selecting the right treatment strategy for a patient, is does not necessarily matter if PFS is compromised as this will not have a huge impact on a disease which persists for 20–30 years. More and more often, we are no approaching treatment for FL the same way we approach DLBCL, HL, and MCL; we should consider FL as a chronic disease and so extending remission duration may not be the best goal for treatment. Friedberg suggested a few potential goals for treatment:

• Changing the natural history of the disease (decreasing transformation, improving survival)
• Remission (making patients feel better; however, active disease is often asymptomatic so may not be an appropriate goal)
• Improving Quality of Life (QoL; especially in symptomatic patients)

Following this, the talk focused on data presented at ASH 2016 (Shadman et al. #616) on the phase III SWOG S0016 trial in newly diagnosed, advanced stage FL patients (NCT00006721), which aimed to compare CHOP combined with rituximab (R-CHOP) or ¹³¹I-tositumomab (CHOP-RIT). After a median follow-up of 9.6 years, estimated 10-year PFS for R-CHOP compared to CHOP-RIT was 42% and 57%, respectively (P = 0.01); so around half of patients are disease free at 10 years. 10-year estimated OS of 82% was achieved with R-CHOP compared to 77% with CHOP-RIT (P = 0.18).

The updated analysis with an extended follow up (median follow-up, 117 months) of the STiL NHL1 study (NCT00991211) comparing bendamustine-rituximab to R-CHOP, presented at the 2017 ASCO Annual Meeting by Mathias J. Rummel and reported on by the LH on 9^th June 2017, was discussed further by Friedberg.  Of the whole patient cohort, including those with MCL, 133 deaths were reported; 65% were attributed to Lymphoma and 35% took place during first remission (Rummel et al. #7501).

Friedberg then outlined a key reason to avoid chemotherapy for FL patients: the risk of secondary malignancies is higher. To demonstrate this, he re-visited the SWOG S0016 trial once again to discuss the incidence of secondary malignancies:

The cumulative incidence of deaths due to AML/MDS was 1.1% (0.2–3.8) with R-CHOP and was 4.1% (2.1–7.1) with CHOP-RIT (P = 0.02; Shadman et al. ASH 2016. #616). The speaker also emphasized that in this study, QoL was not addressed but maybe should have been.

Low tumor burden FL (the majority of patients)

Currently, guidelines state that radiotherapy (XRT) is the standard treatment approach; but, patients with stage 1 FL have been reported to achieve excellent prognoses and equivalent PFS regardless of the treatment choice selected (Friedberg et al. JCO. 2012) and so do not require chemotherapy.

The talk then moved on to discuss whether delaying treatment impacts OS. The Stanford University Experience found that in patients who received treatment immediately (n=688), median OS was 12.1 years compared to 15.1 years in patients who had no initial therapy (n=645). Therefore, there is no evidence to suggest that the timing of initial treatment negatively impacts on OS in patients with FL (Tan et al. Blood. 2013).

“Watch and wait” compared to rituximab was the next topic of the talk. Ardeshna et al. (Lancet Oncol. 2014) conducted a phase III, randomized trial (NCT00112931) comparing these two strategies. After a median follow-up of 3–4 years, those receiving treatment were less likely to progress compared to patients in the observation arm, but OS was the same (3-year OS was approximately 97%). Moreover, 44% of patients in the observation arm have not yet needed treatment and outstanding early outcomes have been reported without chemotherapy. We are currently awaiting the 10-year follow-up data.

Next, the phase III, randomized RESORT trial (NCT00075946) comparing two different dosing regimens of rituximab for low tumor burden patients was discussed:

This trial found no significant difference between the maintenance and re-treatment arms in terms of Time to Treatment Failure (TTF; 94% of patients were alive at 5 years in both arms) or transformation rates. Additionally, 3.5 times more rituximab was administered to patients in the maintenance arm versus those in the scheduled treatment arm, indicating the maintenance patients were being over-treated (Kahl et al. JCO. 2014). It has further been found that no significant different in QoL was found between patients in the maintenance and re-treatment arms; illness-related anxiety was comparable at all time points (P > 0.05), and this, as well as general anxiety, decreased over time in both arms. However, it was noted that patients do like to be in remission as this abates their anxiety (Wagner et al. JCO. 2015).

Friedberg then focused the talk on the phase III, randomized SAKK 35/98 trial (NCT00003280) which aimed to assess the effect of rituximab consolidation in patients with FL who had received induction with rituximab (x4 weekly).

After 10 years of follow-up, it was reported that 45% of chemotherapy-naïve responders were in remission at 8 years. This was described by Friedberg as a “phenomenal result”, and emphasized that very few events occurred after 5 years. Furthermore, there was no evidence to suggest that these patients would have benefited from earlier chemotherapy treatment (Martinelli et al. JCO. 2010).

The phase III, randomized SAKK 35/03 trial, which aimed to compared two different schedules of maintenance rituximab in rituximab-responding patients with untreated, chemotherapy resistant, or relapsed FL (NCT00227695), was then discussed. Overall, 270 patients were enrolled and received rituximab x4 weekly. Of these, 165 responded and were randomly assigned to either 4 more doses of rituximab or 5 years of rituximab. It was found that EFS and OS between the two schedules of rituximab were not significantly different, but increased toxicity resulted from prolonged administration (Taverna et al. JCO. 2016).

Lastly, it has been reported that patients who do not experience early events have outstanding prognoses. In an article by Maurer et al. (Am J Hem. 2016) patients who achieved EFS at 12 months had no added mortality beyond that of the background population. Moreover, baseline FL-IPI was no longer prognostic in patients achieving EFS at 12 months.

Friedberg concluded this portion of the talk by stating that the majority of low tumor burden FL patients do not require chemotherapy.

High tumor burden and high-risk FL

This part of the talk began by presenting results from The Cancer and Leukemia Group B (Alliance) 50401 phase II, randomized trial which aimed to compare rituximab (R), lenalidomide (L), and rituximab plus lenalidomide (LR) in patients with recurrent FL (NCT00238238). Poor accrual led to the discontinuation of the rituximab monotherapy arm.

This trial concluded that the combination of LR is more active than monotherapy with just L in recurrent FL with similar toxicity (incidence of grade 3–4 AEs was 53% and 58%, respectively) and so deserves further investigation for the treatment of B-cell NHL (Leonard et al. JCO. 2015). However, Friedberg emphasized that the LR regimen is a continuous treatment, is expensive, and the low level toxicities (such as rash) can negatively impact a patient’s QoL.

The LR regimen was compared with rituximab plus chemotherapy by Fowler et al. (Lancet Oncol. 2014) as upfront therapy in the phase III, open-label, randomized study RELEVANCE (NCT01476787).

Of the 110 patients enrolled, 50 had FL and 28% of these had high-risk FL-IPI. Of the evaluable patients with FL (n=46), 40 (87%) achieved a Complete Response (CR) and 5 (11%) a Partial Response (PR).

The LYSA group have also tested some strategies such as PD-L1 blockade and direct cell killing in R/R CD20+ NHL. They concluded that treatment should be rituximab-chemotherapy based and some patients will relapse earlier than others. At relapse, do not use chemotherapy, and consider chemotherapy-free options followed by maintenance which may include checkpoint inhibitors.

Friedberg also discussed the US S1608 trial, which is about the open, in R/R grade 1–3a FL patients which aims to compare obinutuzumab combined with either lenalidomide, TGR-1202, or CHOP.

It was designed to answer whether, in the highest risk patients, is chemotherapy or anthracyclines more effective? Some KOLs disagreed with the study design, stating it is wrong to give these patients with chemotherapy but others said it will be fine. This lack of consensus justifies the need for this trial. Restaging will be repeated following 6 cycles of therapy, at 1 year, and at 30 months with PET; the primary endpoint is CR.

The topic of the talk then changed to discuss vitamin D, as lower vitamin D levels at diagnosis of FL has been found to correlate with a poorer outcome after R-CHOP (Kelly et al. JCO. 2015). In their SWOG cohort (newly diagnosed FL patients enrolled onto S9800, S9911, or S0016 between 1998 and 2008), after a median follow-up of 5.4 years, adjusted PFS and OS Hazard Ratios (HRs) were 1.97 (95% CI, 1.10–3.53) and 4.16 (95% CI, 1.66–10.44), respectively, for vitamin D deficient patients (<20ng/mL; 15% of cohort). In their LYSA cohort (newly diagnosed FL patients enrolled onto the PRIMA trial between 2004 and 2007), after a median follow-up of 6.6 years, adjusted PFS and OS HRs were 1.50 (95% CI, 0.93–2.42) and 1.92 (95% CI, 0.72–5.13), respectively, for vitamin D deficient patients (<10ng/mL; 25% of cohort). The ILyAS trail aiming to compare the addition of vitamin D or placebo to standard of care is yet to open. If the addition of vitamin D does make a difference, this has huge implications for the management of all patients treated with anti-CD20 strategies.

The current chemotherapy-free strategies under investigation in FL were listed, and include:

• PI3K inhibitors (an attractive option)
• Venetoclax (activity in FL is modest but the option is available to combine with other agents)
• Ibrutinib (has profound activity in MCL but only modest results in FL)
• Checkpoint inhibitors
• Novel naked and conjugated antibodies
• CAR-T cells and adoptive immunotherapy

Friedberg concluded this portion of the talk with a succinct summary slide:

He went on to say that in a decade’s time, there may still be a group of patients for whom chemotherapy is the best treatment option; however, he did not think it would be a particularly large group of patients. For now, treatment needs to be tailored for each patient:

• Stage 1: XRT (no chemotherapy)
• Low tumor burden: observation, rituximab (no chemotherapy)
• High tumor burden: chemoimmunotherapy, awaiting results from RELEVANCE
• Relapsed disease: dependent upon if shows indolent behavior (rituximab) or “high-risk” relapse (no consensus approach so far)