Highlights in Pathology: Medical Oncology (January 2020)

Dr. Charles Lim, Medical Oncologist, Trillium Health Partners

1. NTRK fusion-positive cancers and TRK inhibitor therapy

JNat Rev Clin Oncol. 2018 Dec;15(12):731-747 [PMID: 30333516] Cocco E, Scaltriti M, Drilon A

This review provides an overview of major advances in the understanding of TRK biology as well as therapeutic targeting of TRK signalling.  

The role of the TRK signalling pathway in neuronal development and differentiation is discussed along with the role of NTRK fusions (NTRK1, NTRK2, NTRK3) as oncogenic drivers.  

NTRK fusion prevalence in adult and pediatric cancers is categorized broadly into common tumour types with low NTRK fusion prevalence versus a group of rare tumours with very high NTRK fusion prevalence.  

Rare cancers enriched for NTRK fusions (frequency > 90%) are highlighted, including:

• mammary analogue secretory carcinoma
• secretory breast carcinoma
• infantile fibrosarcoma
• congenital mesoblastic nephroma.

Clinical applications are then discussed, focusing on trials involving larotrectinib and entrectinib, the tyrosine kinase inhibitors (TKI) furthest along in development at present.  

Three trials of larotrectinib and four trials of entrectinib (phase I or II) enrolling adults and children with multiple tumour types harbouring NTRK fusions are summarized.  

Based on these studies, both agents have received FDA breakthrough designation.  

Response rate in combined analysis of 55 patients enrolled in larotrectinib trials was 75%. Importantly, responses were observed regardless of tumour histology, patient age, or NTRK fusion partner. Similarly high response rates were observed in entrectinib trials.

Off-target adverse events were also less frequent compared to other multi-targeted TKIs, indicating a favourable safety profile.

Mechanisms of drug resistance to first generation TKIs are discussed, and next-generation TKIs currently in development are reviewed. 

The findings from these clinical trials challenge the traditional clinical dogma of tumor-histology driven treatment recommendations.  

The very high response rate combined with favourable toxicity profile is very appealing in clinical oncology, particularly for patients with rare cancers with high prevalence of NTRK fusions.

2. ESMO recommendations on the standard methods to detect NTRK fusions in daily practice and clinical research

Ann Oncol. 2019 Sep 1;30(9):1417-1427 [PMID: 31268127] Marchiò C, Scaltriti M, Ladanyi M, Iafrate AJ, Bibeau F, Dietel M, Hechtman JF, Troiani T, López-Rios F, Douillard JY, Andrè F, Reis-Filho JS

Given the active interest in clinical oncology for NTRK-fusion directed therapies, the ESMO Translational Research and Precision Medicine Working Group released these consensus-based recommendations on methods to identify NTRK fusions to support clinical decisions.  

Multiple methods including IHC, FISH, RNA NGS and targeted-DNA NGS are discussed, along with the major strengths and weaknesses associated with each.

In brief, pan-TRK IHC can be valuable given its high sensitivity/specificity and relatively limited expense, and may be of particular utility in rare tumour types with high NTRK fusion prevalence. However, technical limitations are acknowledged, and an algorithm with pan-TRK IHC used first as a screening tool, followed by confirmatory NGS is proposed.  

FISH is discussed, acknowledging multiple limitations in the NTRK-specific context including the multiple possible fusion partners, the labour-intense nature and the expertise required.  

RNA and DNA NGS are identified as effective methods for NTRK fusion detection in large populations. However, tissue preparation and quality issues are highlighted as limitations, and the inability to identify all actionable NTRK fusions by NGS is acknowledged.  

In cases identified as NTRK fusion positive by NGS, supplementary testing with a second method is suggested. When considering practical applications, efficiency and cost challenges should be considered. The prevalence rate (high vs. low) of NTRK fusions among tumour types is proposed as a means to guide optimal NTRK testing methods. 

Key recommendations from the ESMO Working Group (Figure 2) are as follows:

• In rare tumour types with high known prevalence of NTRK fusions, any validated method is applicable, with FISH and RT-PCR or targeted RNA NGS as the most favourable.  
• In an unselected population of tumour types with low prevalence of NTRK fusions, where NGS is not available, screening IHC with subsequent send-out for NGS is recommended.  
• For an unselected population at centers with NGS capabilities, front-line NGS is preferred if feasible, with RNA NGS considered the gold standard.  
• An alternative two-step approach with initial IHC screening followed by NGS is also proposed. 

Finally, the Working Group recommends a broad approach to selecting patients for testing, casting a wide net to avoid missing patients harbouring NTRK fusions.  

The authors argue “the population to be tested should be represented by any malignancy at an advanced stage, in particular if it has been proven to be wild type for other known genetic alterations tested in routine clinical practice, and especially if diagnosed in young patients.”

3. NTRK fusion detection across multiple assays and 33,997 cases: diagnostic implications and pitfalls

Mod Pathol. 2019 Aug 2 [PMID: 31375766] Solomon JP, Linkov I, Rosado A, Mullaney K, Rosen EY, Frosina D, Jungbluth AA, Zehir A, Benayed R, Drilon A, Hyman DM, Ladanyi M, Sireci AN, Hechtman JF

This article details the experience at Memorial Sloan Kettering from 2014 to 2019 with various assays to detect NTRK fusions.  

Performance of DNA sequencing using MSK-IMPACT panel, RNA sequencing using MSK-Fusion panel and pan-TRK IHC using the EPR17431 clone was compared.  

Detection of NTRK fusion by RNA sequencing defined positive cases. 

87 NTRK fusion positive cases were identified, with various tumour types represented.  In this cohort, NTRK fusions were mutually exclusive with mutations in KRAS, BRAF, NRAS and EGFR. Sensitivity and specificity for DNA sequencing was 81.1% and 99.9% respectively.  

DNA sequencing was not able to identify 14 positive cases, and this was mainly attributed to incomplete intron coverage in the MSK-IMPACT panel (NTRK1, NTRK2 and ETV6 covered).

66 NTRK fusion positive and 317 NTRK fusion negative cases were tested by pan-TRK IHC, yielding overall sensitivity of 87.9% and specificity of 81.1%.  

Sensitivity was lower for NTRK3 fusions and for tumours with neural differentiation or sarcomas.  

Specificity was lower in breast and salivary gland tumours.

Limitations in IHC performance were discussed, with false positives more commonly observed in tumours with neural or smooth muscle differentiation, and false negatives more commonly observed in NTRK3 fusions. 

IHC sensitivity and specificity in sarcoma was low, and thus NGS was recommended instead for this tumour type.  

DNA sequencing technical limitations were also discussed, particularly for rare NTRK fusions.  

The authors conclude that each assay has specific advantages and disadvantages, and the selection of the most appropriate assay should account for tumour type, differentiation and fusion type. From a practical applications perspective, the availability and economic impact of various testing platforms should not be overlooked either.

4. Colonic Adenocarcinomas Harboring NTRK Fusion Genes: A Clinicopathologic and Molecular Genetic Study of 16 Cases and Review of the Literature

Am J Surg Pathol. 2019 Sep 23 [PMID: 31567189] Lasota J, Chłopek M, Lamoureux J, Christiansen J, Kowalik A, Wasąg B, Felisiak-Gołąbek A, Agaimy A, Biernat W, Canzonieri V, Centonze G, Chmielik E, Daum O, Dubová M, Dziuba I, Goertz S, Góźdź S, Guttmejer-Nasierowska A, Haglund C, Hałoń A, Hartmann A, Inaguma S, Iżycka-Świeszewska E, Kaczorowski M, Kita P, Kołos M, Kopczyński J, Michal M, Milione M, Okoń K, Pęksa R, Pyzlak M, Ristimäki A, Ryś J, Szostak B, Szpor J, Szumiło J, Teresiński L, Waloszczyk P, Wejman J, Wesołowski W, Miettinen M

The study screened an unselected cohort of 7008 colon cancer patients using pan-TRK IHC using the A76HR clone, with subsequent confirmation by RNA sequencing.  

All patients underwent partial colectomy and were identified from Europe, Japan and the United States.  

Positive IHC staining was identified in 16 cases (0.23%) with RNA confirmation completed for 15 cases, with 14 cases with NTRK1 and 1 case with NTRK3 fusion.  

Clinical characteristics were reported, with more patients being female (81%), more tumours involving the right colon (75%) and more frequent moderate-poor differentiation and mucinous features on histology.  

13 (81%) cases had deficient MLH1 and PMS2 expression, and 13 (81%) cases had high Ki67 staining > 80%.  

10 cases had sufficient DNA quality for sequencing, and none of these harboured typical colorectal cancer driver mutations (BRAF, KRAS, NRAS). However, 5/9 analyzed tumours had identified mutations in the PI3K-AKT/MTOR pathway.

Although the small cohort size limits the strength of conclusions drawn, the authors highlight the commonalities in phenotypic and genetic factors between colon cancers harbouring NTRK fusions and MMR-deficient colon cancer. However, the authors ultimately conclude that NTRK fusions are a distinct and very rare subgroup of colon cancers.