Hyperprogression and Pseudoprogression - Faster than the Speed of Light
Immunotherapy has revolutionised the treatment of multiple cancers over the past fifteen years, many of which were invariably fatal with conventional cytotoxic agents. At the same time, targeted therapies continue to advance a personalised approach to precision oncology with the directive to turn cancer into a chronic disease. Despite these advances, clinical challenges remain, including sequencing of therapy, pseudoprogression, hyperprogression and other drug-related complications. This review will focus on the diagnostic dilemma of differentiating pseudoprogression (PSP) from progressive disease (PD) and hyperprogression (HPD).
The first and perhaps biggest problem is the consensus on the definition between PSP and HSP, or lack of one. The consequences of these three distinct entities are obvious; PSP implies immune checkpoint inhibitor (ICI) efficacy, and the worsening radiological appearance of cancer can lead to premature treatment discontinuation. HPD, as the name suggests, is the exact opposite: true and rampant cancer progression that usually leads to exceptionally poor patient outcomes. One example is in the head and neck space, where patients who experienced HPD had progression-free survival (PFS) of 2.5 months, compared to 3.4 months in those who did not.
PSP was first described in patients receiving temozolomide for brain cancer. In modern practice, however, it is more frequently observed in patients receiving ICI as part of systemic therapy. Depending on tumour type, the incidence of pseudoprogression with immunotherapy ranges from 3.7-15%. The underlying mechanism is believed to be related to a rapid influx of immune cells in the tumour bed. Radiological and pathological changes consistent with PSP are characterised by oedema and immune cell necrosis surrounding the tumour. These changes usually occur in the first 12 weeks of treatment. ICI-causing PSP is associated with higher rates of survival at 12 months compared to true PD.
The reported incidence of HPD varies from 5.9-43% across various studies. The mechanism of hyperprogression is unclear, and it is unknown if this is due to the biology of the tumour or the specific ICI. A study by Kato S et al. defined hyperprogression as time to treatment failure <2 months, >50% increase in the tumour burden, and a two-times fold increase in the growth rate of cancer. Concurrently Lo Russo G's study required 3/5 criteria to diagnose HPD: time to treatment failure <2 months, >50% in the diameter of the target lesions, 2x new lesions in the affected organ, new lesions in a new organ and decline in performance status. Matos et al. required a >40% increase in the sum of target lesions or new lesions in at least two other organs. However, one consistent limitation of these definitions is that most did not provide reference data on the tumour growth rate before ICI administration.
Risk factors for those at risk of hyperprogression include older age, female and performance status of 1-2. As mentioned previously, treatment cessation is the best way to stop hyperprogression and identifying an adequate replacement is required. Still, despite so many advances in treatment and diagnosis, many questions remain on how best to identify those with true hyperprogression while ensuring those that respond continue the optimal treatment regimen. Most clinical studies for HPD are retrospective, and controversy regarding whether this is a natural or accelerated growth process due to ICI exist. The author of this review suggests ongoing MDT input utilising specialists from various fields, including experts for the different cancers with experience in that specific treatment, radiologists, and pathologists to ensure a coordinated approach putting the patient's health and safety as the primary objective. Consensus is necessary to help facilitate diagnosis and provide a standardised approach to diagnosis and treatment of hyperprogression.
Trends in Cancer- Hyperprogression and immunotherapy: Fact, Fiction, or Alternative Fact?
CT and MRI are the two most common imaging modalities in the diagnosis and monitoring of solid organ malignancies; however, in some cases, they are the root cause of confusion between PSP and HPD. A study by Topkan et al. noted that 12 out of 28 patients with confirmed pseudoprogression had unnecessary surgery due to misclassification on MRI. PET scans have been demonstrated to identify early pseudoprogression in glioma patients receiving chemoradiotherapy. In addition, a small retrospective of patients with melanoma brain metastases on immunotherapy series found that low tracer uptake was associated with pseudoprogression. However, significant unknowns remain, and further studies in this area are required.
Biopsies of sites of potential progression are critical in differentiating PSP from HPD. Pathological characteristics of PSP include dense CD8+, granzyme B+ lymphoid infiltrate and TIA1+. Infiltration of normal lymphocytes, such as CD3, CD4 and CD8, rather than tumour cells may also be present. However, the need for a definitive diagnosis must be balanced by the risks inherent to an invasive procedure, and the sites of disease may further limit an interventionalist’s ability to gain an adequate sample.
A surrogate marker that is growing in prominence in differentiating PSP from HPD is ctDNA (circulating tumour DNA). It is currently hypothesised that undetectable ctDNA at baseline or a decrease from the baseline level could predict pseudoprogression. This was illustrated in a case report by Guilbert et al. that followed nine pseudoprogression patients with ctDNA. The study found that ctDNA patients with melanoma treated with PD-1 inhibitors could predict pseudoprogression with a sensitivity of 90% and specificity of 100%. More extensive studies with more cancer types are needed to prove this relationship, but ctDNA may yet prove to be a less invasive method of differentiating these two diametrically opposite clinical entities.
The Bottom Line: pseudoprogression and hyperprogression are two rare and important clinical entities. Both are rare and associated mostly with immune checkpoint inhibitor therapy, but their impact on patient outcomes can be significant. Current issues with the definition of PSP and HPD limit prompt differentiation on the basis of radiology alone, and further investigations, such as direct biopsy or, more recently, ctDNA assays, may be required to differentiate.
References
3. https://www.frontiersin.org/articles/10.3389/fphar.2021.678409/full
6. https://www.cell.com/trends/cancer/fulltext/S2405-8033%2820%2930018-2
