ROS1 Rearrangement: Molecular Oncology Tumor Boards

ASCO University
Feb 11, 2015 9:33 AM

User Instructions: Welcome to the Molecular Oncology Tumor Board Series! This educational initiative is a collaboration between the American Society of Clinical Oncology (ASCO), College of American Pathologists (CAP), and Association for Molecular Pathology (AMP).

A new case will be presented each month with discussions led by an expert pathologist and medical oncologist. This month’s topic is led by Drs. Jamie Chaft (Memorial Sloan Kettering Cancer Center) and Laura Tafe (Dartmouth).

Users are encouraged to leave comments and post questions about the case in order to generate a wide discussion among the cancer care community. You can also receive email notifications when new comments are posted by clicking the “Email me when this thread has replies.” option located at the bottom of this page.

When posting, please abide by the terms and conditions of this website.

Please click here for supplemental resources related to the case.



ASCO University
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 11, 2015 9:39 AM

Patient Case

Age/Sex: 40 year-old female
Medical History: This is a 40 year old woman never-smoker who was well until she developed cough and fever.  She initially thought her symptoms were infectious, though there was no resolution and instead she developed night sweats and weight loss.  She ultimately presented with right-sided chest pain and imaging performed revealed a left lower lobe nodule, liver metastases, and extensive multisite lymphadenopathy.  At the time of medical oncology consultation she remained symptomatic and had interval development of dyspnea and rare bright red hemoptysis.

Co-morbidities: None

Pathology: A CT guided biopsy of the liver revealed metastatic, poorly differentiated adenocarcinoma consistent with a lung primary. 
Immunohistochemical stains support a lung primary:
Positive – CK7, TTF-1, Napsin A
Negative – GATA-3, ER, PR, CDX-2, CK20
Initial molecular studies: Immunostains for EGFR L858R and ALK were negative.  PCR-based sizing assay to detect EGFR Exon 19 deletions was negative and further molecular diagnostic testing was requested. 
Staging evaluations: Completion of staging revealed bone metastases on PET.  Brain MRI was negative for metastatic disease. Representative images are demonstrated here: Please click here.


ASCO University
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 11, 2015 9:41 AM

Discussion Questions

  1. What type of molecular testing would you request?

  2. Would you await the results of molecular diagnostic studies or initiate therapy promptly?

  3. What systemic therapy would you offer?


Benjamin Philip Levy, MD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 11, 2015 2:07 PM

Given her symptoms, I would be inclined to institute systemic chemotherapy. Her unsuccessful preliminary search for actionable mutations (EGFR and ALK) warrants additional testing via next generation sequencing (NGS). Many alterations to be on the lookout for, but would be interested in hearing from others what they would test for and whether they would send off for particular alterations or just send NGS to get the full panel. 


Anis Toumeh, MD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 11, 2015 3:02 PM

This is clearly a case where there is a high chance of detecting a genetic alteration giving the young age and non-smoking status. 

Unfortunately, this patient's tumor doesn't harbor sensitizing mutations in the EGFR gene nor an alteration in ALK for which we have established very effective targeted therapies.

The two recent reports by Mazieres et al (JCO). and Shaw et al. (NEJM) document the activity of Crizotinib in tumors that harbor rearrangements in the ROS1 gene regardless of the partner gene making testing for ROS1 rearrangements attractive in her case despite that only 1-3% of NSCLC harbor such alterations.

Moreover, testing for the BRAF V600E mutation (2% of NSCLC) might be worthwhile as well giving the activity of BRAF inhibitor Dabrafenib in such tumors as shown in a phase II open label trial presented in ESMO 2014. 

For this particular case, I would consider NGS to look for genetic alterations that would allow her to participate in a clinical trial. 

Meanwhile, I would start her on combination platinum based chemotherapy in the form of Cisplatinum and Pemetrexed in addition to involving palliative care early in the disease course awaiting the NGS results.  


Nathan A. Pennell, MD, PhD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 11, 2015 4:43 PM

I agree with the prior commenters, this patient needs urgent systemic therapy and the best available choice given her lack of EGFR or ALK alterations is platinum doublet chemotherapy. I would use carboplatin and pemetrexed in this situation.

Further molecular testing is certainly warranted in a young never smoker, either with a hotspot panel to include the most common molecular derangements in NSCLC, or with a broader NGS panel including the same list. I would think including ROS1 and RET gene fusions (by FISH or NGS) along with MET amplification would be potentially helpful, along with BRAF and HER2 mutation testing which should cover most of the currently actionable alterations outside of a clinical trial.

One question I would have for the expert discussant is whether we have reached a point where broader testing (beyond EGFR and ALK) should be routinely done at diagnosis, or should we wait for initial testing to come back negative? And would you test all patients or just the never/light/distant smokers who might be enriched for the actionable alterations?


Nathan A. Pennell, MD, PhD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 11, 2015 5:14 PM

I'd also like to hear people's opinion about whether even EGFR and/or ALK are being routinely tested in all new non-squamous patients outside of academic centers? I hear mixed things: some test routinely but many only do it on request when they think it warranted.


Michael A. Thompson, MD, PhD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 12, 2015 7:35 PM

If I got lung cancer or a family member I would suggest broad actionable target testing.
Our cytotoxic traditional chemo just isn't that good.
Molecularly targeted therapy doesn't appear to be curing people, but it provides more options for treatment and potentially less toxic and more convenient.



Michael A. Thompson, MD, PhD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 12, 2015 7:37 PM

It should be.
It isn't.

And now we have blood based molecular testing as a capability for frail, elderly that have COPD and conern about repeat biopsy when tissue limited.



Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 13, 2015 3:06 PM

It is difficult to know what is universally being tested in the community.
 I work in a community setting and check EGFR ALK and ROS -1 routinely on stage 4 non squamous histology.  I selectively use NGS panels such as Foundation One.  
Patients presented at our tumor board seem to be approached in a similar fashion by the physicians who participate.  
 I have a colleague (Fadi Braiteh MD) who has thought about how to improve Molecular testing in the community setting.  if pathology reflexively tests all samples that will lead to unnecessary testing and increased cost.  His thoughts are a comment in bold on the report such as "molecular testing for EGFR mutations ALK rearrangements and ROS-1 rearrangements is available on request" could work as a reminder. Or use of a navigator, and a semi automatic testing paradigm, where the pathology report is forwarded to the nurse navigator who can check the stage and if the test has been performed on any prior specimens and then release the order to perform molecular testing if appropriate and/or discuss with the treating physician before the appointment.
If there are concerns that testing is not being performed consistently in the community these systems might help.
By the way I am loving this forum really enjoyed the discussion so far, and the previous colorectal case also excellent.


Craig Uthe, MBA, BS
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 13, 2015 4:41 PM

As an online patient advocate of targeted therapies (and crizotinib-for-ROS1+ patient for 3+ years so far), I would not want to see a peer patient with a fair chance of ROS1+ (or other experimentally-useful driver mutation/rearrangement) miss the opportunity to find out if they had something potentially useful.  Some background thoughts and references for your discussion:=== The Odds ===The odds of a never-smoker (like the case here) having an at least experimentally-promising driving mutation have been said to be better than 50% and "almost all" never-smokers if you include experimental treatments that might not turn out as well as hoped.  See Figure 1 summarization of prior data shown in this 2013 article and the text referring to it: that if one has ruled out common driving mutations, the odds of having one of the remaining otherwise-rare ones increases significantly, including experimentally useful ones.  Picture that pie chart reformed without the EGFR or ALK wedges (and then maybe without the KRAS wedge which remains big until ruled out).  FWIW#1, being relatively young might further increase the odds of certain promising or experimentally-useful mutations/rearrangements, per research studies.  I could be mistaken but I'm under the impression that gender usually isn't significant (except maybe for EGFR in never-smoker women in Asia) even though I have a vague recollection that there might have been slightly more women with a given useful mutation than men in some mutation-specific studies I read a couple of years ago.FWIW#2, am I mistaken or did some reports suggest if the patient's cancer cells had solid signet-ring cell or acinar/mucinous cribriform pattern their odds of ROS1+ or RET+ might improve further?:
although this landmark paper introducing ROS1 NSCLC did not find signet ring in ROS1 patients: Not Always "Useful" ===I believe that molecular test results, even if not medically useful, often help patients psychologically by helping them understand the nature of their mysterious disease a little.  In my own case, I sought further "experimental" molecular testing even though the odds of medical usefulness at the time seemed close to zero -- I wanted to know the name of my enemy, its nature, and wanted to follow emerging research on it.  Even a "not identifiable/not known" result would've been worth knowing and worth a try.  (As it turned out, though, I had one of those unproven rare mutations and became one of the first dozen to benefit from a drug targeting it.)P.S. -- Rare cancer drivers might lead to experimental trials rather than approved-use or off-label Rx, hopefully trials already showing some success.  Although it doesn't seem applicable to the case presented here, it's worth acknowledging that some patients are not well enough or cannot afford to travel to trial locations for promising experimental treatments that might apply to them, or who would not meet eligibility criteria of any promising trial.  Such limitations reduce the medical usefulness of test results where the applicable treatment would not be available locally via Rx, local trial, or compassionate use application in time.


Jamie E. Chaft, MD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 16, 2015 11:45 AM

Course Faculty Response

There have been some great thoughts, questions and issues raised in this discussion.  What seems to be the topic of most interest is molecular testing – who, what and why?  

Who - From my perspective, while clinical characteristics enrich for detection of certain molecular aberrations, the actionable mutations also occur in patients without those characteristics (D’Angelo, S. J Clin Oncol. 2011).  Therefore, all patients with non-squamous NSCLC, adenosquamous carcinoma or a small biopsy demonstrating squamous cell carcinoma without the typical smoking history to correlate should be considered for testing.  

What – The what to order is a bit tricky with the underlying issue being tissue availability.  While the ASCO Guidelines endorse EGFR and ALK testing in all patients, extracting DNA for EGFR only and using tissue for ALK FISH will leave many potentially actionable oncogene-driver mutations in question and potentially a shortage of tissue.  Multi-signal FISH probes are in development for the investigation of multiple gene rearrangements in a single tissue section, though these are not routinely utilized.  Next-generation sequencing platforms are available at many academic centers and commercially.  If there is sufficient tissue for NGS, this avoids the issues of tissue consumption that may arise from a stepwise testing strategy, evaluating for point mutations, deletions, insertion, rearrangements and amplifications at the same time with limited tissue requirement.

Why - The why question is beautifully answered in the previous comment by Craig Uthe, patient and online patient advocate.  To bolster his perspective and for references relevant to other targets, I would direct anyone interested in more details to the NCCN Guidelines, version 1.2015 on Non-small cell lung cancer where a table on emerging targeted agents for patients with genetic alterations outlines actionable oncogene targets and available drugs, beyond EGFR and ALK.


Laura Tafe
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 17, 2015 9:26 AM

What type of molecular testing would you request?

Course Faculty Response

Molecular testing guidelines put forth by CAP/IASLC/AMP for the selection of lung cancer patients for EGFR and ALK tyrosine kinase inhibitors were released in July 2013 (Lindeman NI, J Thorac Oncol 2013 Jul;8:823-59). This consists of 37 evidence-based recommendations to address, among other issues, who should be tested, when should molecular testing occur and how should EGFR and ALK testing be performed.  However, it is recognized that in the meantime a number of other important and targetable gene alterations have emerged such as ROS1 and RET rearrangements and currently the guidelines are under revision. Many institutions routinely perform additional molecular testing beyond EGFR and ALK analysis.   
In this patient, initial testing for EGFR Exon 21 L858R mutation and ALK rearrangement by IHC were negative.  These are both highly specific but imperfectly sensitive assays.  EGFR Exon 19 deletions were assessed by a PCR-based sizing assay which is rapid and both highly sensitive and specific.  With no EGFR or ALK mutations identified, the test of choice in this young, never-smoker with limited available tumor tissue for analysis was a multiplexed next-generation sequencing (NGS) panel.  

Methods: MSK-IMPACT (Integrated Mutation Profiling of Actionable Cancer Targets) was used to identify specific mutations in 341 genes.  This is hybridization capture-based next generation sequencing (NGS) assay run on an Illumina HiSeq 2500 instrument.  This assay is able to detect single nucleotide variants (SNVs – eg. point mutations), insertions and deletions, structural variants (eg. rearrangements/translocations), and copy number variants (CNVs).  Data analysis is performed using a custom pipeline.

Tissue:  FFPE tissue is first reviewed by a pathologist and the area of tumor is identified and subsequently macrodissected to ensure >50% gross tumor content.

DNA:  DNA is extracted from the FFPE using a Qiagen DNA FFPE kit. 

Mutation assessment: Mutations are called against the patient's matched normal sample (blood). This assay reports somatic variants confirmed to be absent in the matched normal (thus ruling out germline variants).

Copy number assessment: The criteria for copy number gain (amplification) and deletions are as follows: if the fold change is greater than 2, it is reported as a copy number gain (amplification). If the fold change is -2 or below, it is reported as a deletion. 

The table attached here lists the exons in the genes that have been clinically validated.  In addition, the remaining genes are considered the “investigational panel”.  


Jamie E. Chaft, MD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 17, 2015 9:39 AM

Would you await the results of molecular diagnostic studies or initiate therapy promptly?

Course Faculty Response

This is a symptomatic patient with widely metastatic lung cancer that presented over a fairly short time period and with poor prognostic signs such as weight loss.  Despite her never-smoker status, it is possible that no actionable oncogenes will be identified and preliminary tests to assess for the common oncogenes are negative.  In this case, it is clinically appropriate to initiate systemic chemotherapy promptly.

Further support of systemic chemotherapy is the lack of overall survival advantage proven from first-line use of a targeted therapy, if an actionable target is discovered.  While tyrosine kinase inhibitors (TKI) are clearly superior in terms of response rate, they are also effective in the 2nd line setting and appropriate selection of a TKI is essential.  The only study to demonstrate an overall survival advantage to first-line use of a TKI was the recently presented pooled analyses from LUX-Lung 3 and LUX-Lung 6 which demonstrated an overall survival advantage with the use of first-line afatinib over chemotherapy in patients with EGFR Exon 19 deletions.  This analysis is flawed as an unplanned pooled analyses and a large cohort of patients randomized to chemotherapy that did not receive a 2nd or later line TKI.  Furthermore, the results of the EGFR fragment analyses confirm that this patient’s tumor does not harbor an EGFR Exon 19 deletion.


Jamie E. Chaft, MD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 17, 2015 9:43 AM

What systemic therapy would you offer?

Course Faculty Response

The choice of first-line therapy is in some ways an art as well as a science and the NCCN guidelines provide many treatment options.  The treatment decision for a patient with non-squamous lung cancer who was cisplatin eligible was once straight forward, with cisplatin + pemetrexed demonstrating clear superiority over cisplatin + gemcitabine (Scagliotti GV.  J Clin Oncol. 2008;26:3543-51).  The selection of agents was further solidified by the demonstration initially of a disease free survival advantage and later by an overall survival advantage of following first-line cisplatin and pemetrexed with maintenance pemetrexed in the PARAMOUNT study (Paz-Ares LG. J Clin Oncol. 2013;31(23):2895-902).  In this patient, who is ineligible for bevacizumab due to hemoptysis, cisplatin and pemetrexed is the regimen with the strongest evidence.

This selection would be much less clear if the patient were eligible for bevacizumab or not cisplatin eligible.  The AVAPERL trial evaluated cisplatin + pemetrexed + bevacizumab and randomized at the time of disease stability to pemetrexed + bevacizumab versus bevacizumab alone (Barlesi F. Ann Oncol. 2014;25(5):1044-52).  This showed a clear superiority in progression free survival in the combination arm but was underpowered to show an overall survival difference, despite numeric superiority in the combination arm, therefore, the approach has been subject to criticism.  Admittedly, I prefer combination maintenance therapy despite a robustly proven overall survival advantage.
For patients who are not cisplatin eligible, the POINTBREAK study compared carboplatin + pemetrexed + bevacizumab followed by combination maintenance to carboplatin + paclitaxel + bevacizumab followed by bevacizumab maintenance.  This study showed a small advantage in progression free survival of the pemetrexed based induction therapy but no difference in overall survival between the study arms (Patel J. J Clin Oncol. 2013;31(34):4349-57).  For a cisplatin-ineligible patient, these are both reasonable options and selection can be dictated by toxicities (more fatigue in the pemetrexed arm, alopecia and neuropathy related to paclitaxel).  For cisplatin-eligible patients, comparison to a carboplatin-based regimen has never been done in a head to head fashion, therefore defaulting to a carboplatin + paclitaxel + bevacizumab is not appropriate in all situations.


ASCO University
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 17, 2015 9:46 AM

Patient Case Update
Patient is treated with cisplatin and pemetrexed.  She has symptomatic improvement after 2 cycles but CT scan shows mixed response with improvement in the lung and some of the liver metastases but progression in other liver lesions and new bone metastases.
Next generation sequencing results are now available. 
Patient results

Negative for mutations in the clinical validated panel.
Positive for ROS1 (6q22) rearrangement in the investigational panel.


ASCO University
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 17, 2015 9:48 AM

Discussion Questions

  1. As the next-generation sequencing results reveal a ROS1 rearrangement in the investigational panel, is there a confirmatory test that should be performed?
  2. The patient had a mixed response to first-line chemotherapy, what would be your treatment recommendation?


Anis Toumeh, MD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 17, 2015 10:09 AM

I would start Crizotinib in this patient in place of chemotherapy. Not sure if we need a confirmatory test


Alexander E. Drilon, MD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 17, 2015 11:23 AM

I agree with starting crizotinib at this point if possible. The drug is clearly active in ROS1-rearranged lung cancers with data from the phase I expansion cohort published in the NEJM (Shaw et al, Crizotinib in ROS1-Rearranged Non-Small Cell Lung Cancer, November 2014). Objective response rate was 72% with a median PFS of 19.2 months (patients still in follow-up). 

While crizotinib is not yet FDA-approved for the treatment of ROS1-rearranged lung cancers, it is listed in the current NCCN guidelines as an emerging targeted agent for patients whose lung cancers harbor this fusion. We have had success securing the drug in clinic and responses can be dramatic and durable. A clinical trial of a ROS1-directed TKI is likewise reasonable, and several drugs are currently in clinical testing. In the absence of a ROS1-directed option, I would entertain switching to alternate chemotherapy.

A variety of methods can be used to test for ROS1 rearrangements including RT-PCR, FISH, and next-generation sequencing. ROS1 IHC is also being explored. Which of these methods will evolve to become the gold standard for testing remains to be seen, however, next generation sequencing has been adopted as the primary screening method for ROS1 fusions in many centers. Each platform will certainly require validation.

If the ROS1 rearrangement detected in this case was previously described (i.e. known upstream fusion partner, preclinical/clinical data on sensitivity available) and the patient has a clear response to crizotinib, in my practice, I would not perform a second test for ROS1. I would consider doing FISH if a novel or complex fusion were detected, or if the patient had a suboptimal response or no response to ROS1 tyrosine kinase inhibition.


Nathan A. Pennell, MD, PhD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 19, 2015 5:33 PM

I also agree with crizotinib in the setting of progressive disease, given the limited but extremely promising data from Alice Shaw's study.

How about I pose a different scenario? What if the patient did NOT progress and had responding or stable disease on chemotherapy, and completed 4 cycles without significant toxicity? Would you proceed with pemetrexed maintenance, knowing the ROS1 was positive? Or switch to crizotinib anyway?

I know my answer but would like to see what others think! This situation comes up not infrequently when patients urgently start chemo and then I find out they have an EGFR mutation or ALK translocation after 1 or 2 cycles.


Jamie E. Chaft, MD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 20, 2015 9:41 AM

Course Faculty Response

Dr. Pennell raises a great question – what to offer in the setting of a newly detected oncogene driver mutation while disease is responding or controlled with standard cytotoxic chemotherapy?  

Here I'll also take the opportunity to make the point that the next generation sequencing in the presented case scenario could have also identified the more common EGFR L858R mutation, as the screening IHC used as a rapid 'first-pass' diagnostic is specific but not highly sensitive.  In my eyes, whether an EGFR L858R with its FDA-approved agents erlotinib/afatinib or a ROS1 with the off-label option of crizotinib were detected, in the setting of disease control and reasonable drug tolerance I would recommend the patient continue chemotherapy.  This allows the TKI to be utilized at the first sign of progression or possible clinical trial enrollment at that time.


Michael A. Thompson, MD, PhD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 20, 2015 9:46 AM

Great comments.
I wish I could upvote/like.

With non-curative therapies -- including targetted therapies -- it is a marathon, not a sprint.
More treatment options including lower toxicity allow a greater quiver of arrows to shoot at targets, but we aren't smart enough to sequence or combine them (yet?) to cure.



Laura Tafe
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 23, 2015 10:06 AM

As the next-generation sequencing results reveal a ROS1 rearrangement in the investigational panel, is there a confirmatory test that should be performed?

Course Faculty Response

Because the investigational panel has not been clinically validated, an alternative test to confirm the ROS1 rearrangement is indicated.  Perhaps the most useful clinical assay to confirm the ROS1 rearrangement is a FISH assay.  Because of the FISH probe design, a ROS1 break-apart FISH assay would be able to detect a rearrangement in ROS1 without needing to know the partner gene.  A single FISH assay utilizes very little tissue and can be performed in most laboratories.  FISH may be used as the primary test to detect a ROS1 rearrangement or it can be a confirmatory test after screening with IHC or to confirm NGS results.

Click here for FISH image:

Like ALK, ROS1 can undergo rearrangement with multiple different partner genes including CD74, SDC4, EZR, SLC34A2 and TPM3 resulting in multiple different transcripts and fusion proteins (Takeuchi K, Nat Med 2012;18:378-381).  Reverse transcriptase (RT) PCR assays require knowledge of the transcript sequence(s) you are trying to detect in order to design the primers.  In this case, where NGS has been performed and the partner gene would likely be known, RT-PCR designed to identify that specific transcript would be another option for confirming the presence of the ROS1 rearrangement.  However, RT-PCR is not an ideal screening assay due to the large number of partner genes and possible rearrangements.

Similar to the ALK IHC that was used in this case, there are also ROS1 antibodies for IHC applications.  ROS1 mRNA is overexpressed in 20-30% of lung adenocarcinoma, and initial available ROS1 antibodies were not effective in identifying rearrangements.  Recently, a ROS1 D4D6 antibody was developed that is much more reliable at identifying ROS1 rearranged tumors.  Several studies have evaluated this antibody and found that it may have utility principally as a screen for ROS1 rearrangement as it is highly sensitive when compared to fluorescence in situ hybridization (FISH) but, with a lower specificity. (Rimkunas VM, Clin Cancer Res 2012;18:4449-4457; Sholl LM, Am J Surg Pathol 2013;37:1441-1449; Mescam-Mancini L, Lung Cancer 2014;83:168-173.)  IHC assays for molecular targets have been incorporated into pathology practice and typically can work well as screening tools but, because of false positive rates, may require confirmatory testing to avoid inappropriate treatment.   


Jamie E. Chaft, MD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 23, 2015 10:08 AM

The patient had a mixed response to first-line chemotherapy, what would be your treatment recommendation?

Course Faculty Response

As the patient has not uniformly responded to cytotoxic chemotherapy, a change in systemic therapy is appropriate.  A clinical trial would be the best treatment option.  If a clinical trial is not available, I would consider prescribing crizotinib.  While crizotinib is not FDA-approved for ROS1-rearranged lung cancer, it is listed in the NCCN guidelines as an appropriate therapeutic agent for this ‘emerging target’.  Based on data published late last year, crizotinib at the FDA-approved dose of 250 mg by mouth twice a day had marked efficacy in ROS1-rearranged lung cancer (Shaw AT. NEJM. 2014;37(21):1963-71).  In this phase 1 expansion cohort, tumors were identified using FISH for ROS1 fusions, though a single patient was identified using RT-PCR.  The objective response rate was 72% and median PFS was 19 months.  As this therapy is well tolerated and early data demonstrate response far superior to that expected from second line chemotherapy, crizotinib is a rational second line treatment option.


Laura Tafe
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 26, 2015 9:15 AM

Course Faculty Summary

Thanks to everyone who participated in this discussion. In summary, these are my individual interpretations based off of the available information:

  1. Next-generation sequencing (NGS) is rapidly becoming incorporated into routine laboratory testing.  NGS testing in tumors has significant advantages over multiple single gene assays.  NGS assays are multiplexed and multiple patients can be tested simultaneously using a relatively small amount of tumor tissue at a decreasing cost.  
  2. Not all NGS assays are equivalent.  Depending on the technology and methodology there are differences with respect to number of genes tested (“hotspot” panels, exome panels, etc.) and the ability to detect various types of genetic variants (single nucleotide variants (SNVs) insertions and deletions, structural variants, and copy number variants (CNVs)).  Therefore, it is essential to know what your laboratory’s assay can and cannot detect.
  3. ROS1 rearrangements are present in approximately 1-2% of NSCLCs, more commonly in patients who were never or light smokers and have tumors with adenocarcinoma histology. Testing should be considered in all non-squamous NSCLC (including adenosquamous carcinoma and mixed adeno-small cell carcinoma) and small biopsies showing squamous morphology but lacking the typical smoking history.
  4. Similar to ALK rearrangements, ROS1 has multiple possible fusion partner genes including CD74, EZR, SDC4, SKC34A2 LIMA1, MSN, and TPM3.
  5. ROS1 break-apart FISH successfully identifies ROS1 rearranged NSCLC and was the primary method used the phase 1 crizotinib clinical trial (Shaw AT. NEJM. 2014;37(21):1963-71). Rearrangement detected in >15% of tumor cells is considered a positive result.
  6. Additional molecular techniques such as NGS and RT-PCR assays can also identify ROS1 rearrangements.
  7. Using a clinically validated laboratory test to make treatment decisions is essential.  ROS1 break-apart FISH can be performed in most laboratories and can be used to screen NSCLC cases for rearrangements and also to confirm rearrangements that may be identified in research-based testing.
  8. Reimbursement for molecular diagnostic tests is highly variable and remains a challenge for laboratories.


Jamie E. Chaft, MD
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 26, 2015 9:21 AM

Course Faculty Summary

Thanks to all for a great discussion.  Based on the case and the topical direction of the question and comments, below are my individual case summary points of interest:

  1. When a patient is unwell, no matter how high the pre-test probability of finding a targeted treatment option, starting chemotherapy while ordering/awaiting results of molecular testing is never the wrong answer.
  2. Don’t forget about cisplatin.  Our supportive medications have come a long way. Particularly in a bevacizumab ineligible patient.  Cisplatin + pemetrexed is better than cisplatin + gemcitabine for lung adenocarcinoma and NSCLC, NOS. In a patient who can receive bevacizumab, I would recommend cisplatin + pemetrexed + bevacizumab though admittedly the question of appropriate regimen and maintenance has not been clearly answered and others may disagree.
  3. Molecular testing is advancing faster than any other aspect of cancer diagnostics and therapeutics.  Arguably, it is advancing faster than the professional society guidelines can keep pace with. A few general rules of thumb about molecular testing from a clinician’s perspective:
    1. Per guidelines, all patients with lung adenocarcinoma, adenosquamous carcinoma, large cell carcinoma or NSCLC, NOS should have their tumor tissue tested for mutations in EGFR and rearrangements in ALK.  If the diagnosis is squamous cell carcinoma made on a small biopsy, testing should be considered.
    2. If immunohistochemistry is used to screen for these aberrations, confirmatory genomic testing should be done – IHC is specific but not sensitive for EGFR or ALK, and it is not specific for ROS1, therefore at present should not be used to consider a tumor ROS1
    3. Oncogene driver mutations are mutually exclusion – therefore if tumor is known to be KRAS+, testing for other oncogenes is not necessary. 
    4. There are many other ‘actionable’ oncogenes in lung cancer – in the setting of a clinical trial or with off-label use of tyrosine kinase inhibitors approved for the treatment of other diseases.  Next generation sequencing looks for EGFR, ALK and the other known aberrations with a relatively low requirement for tumor tissue. NGS should be considered and/or discussed with the patient.
    5. Interpreting NGS results can be confusing. If you’re not entirely clear, be sure to ask someone very familiar with the lingo before prescribing a TKI.
  4. ROS1 is one of many actionable oncogenes that may be identified with NGS with very promising early clinical data available.
  5. There is no rationale to add a targeted therapy to an ongoing effective chemotherapy.  There is also no great reason to suggest switch from an effective chemotherapy to a TKI in the absence of toxicity or patient preference.  As highlighted in the discussion, the TKI can serve as a next line of therapy.  
  6. For patients with lung adenocarcinoma, adenosquamous carcinoma, large cell carcinoma or NSCLC, NOS who do not have sufficient tissue available for molecular testing and in whom a tumor biopsy is not contraindication, consideration should be given to repeat biopsy for testing.  Though, stay tuned, as blood-based testing for mutations on circulating tumor DNA will likely make its way into the Molecular Oncology Tumor Boards in the near future.


ASCO University
Re: ROS1 Rearrangement: Molecular Oncology Tumor Boards
Feb 26, 2015 9:27 AM

Thank you to Drs. Chaft and Tafe for leading the discussion of this case and also to all of those who contributed to the conversation! The forum is now closed to further comments but users have the opportunity to claim CME credit on ASCO University by accessing this link.

Please check back in mid-March for a new case in this series related to Breast Cancer.