*26 Abrams, J.

In May, 1986, Bernadette M. Policelli died of breast cancer. In August, 1986, the plaintiffs, as executor and executrix of Policelli’s estate, were substituted as plaintiffs in a complaint against the defendant, Dr. Armen Roupenian. In October, 1986, a jury returned a verdict for the defendant. Following the verdict, the plaintiffs filed a motion for a new trial which was denied. The plaintiffs appealed from the judgment and from the denial of their motion for a new trial. On appeal, the plaintiffs asserted that the trial judge erred in excluding the opinion testimony of their expert witness, the decedent’s treating oncologist. The plaintiffs argue that the exclusion of the expert’s testimony was prejudicial error. In an unpublished order and memorandum, the Appeals Court affirmed the judgment and the denial of the motion for new trial. 28 Mass. App. Ct. 1108 (1990). We allowed the plaintiffs’ application for further appellate review. We agree with the plaintiffs that it was prejudicial error to exclude the opinion of the plaintiffs’ expert. We reverse and remand for a new trial. 2

Bernadette Policelli consulted her family physician on December 24, 1981, concerning a lump she had discovered in her right breast. The physician palpated the lump in the upper lateral quadrant of her right breast, and recorded the location of the lump he felt by making a schematic drawing in his office record. The physician thought that the lump was probably benign, but he was not entirely comfortable with his diagnosis and so arranged for a mammogram and a surgical consultation with Dr. Roupenian. On December 30, 1981, another physician, a radiologist, performed a mammogram. The mammogram revealed “an ill defined density” above the nipple of the right breast, but was not diagnostic of any disease. The radiologist also performed an ultrasound examination, which did not demonstrate a purely benign, cystic condition. The radiologist felt that a biopsy was indicated and had someone from his office call Policelli’s doctor to make *27 sure further action was taken to diagnose Policelli’s condition.

Policelli then went to Dr. Roupenian on January 5, 1982. Dr. Roupenian palpated a lump in her right breast, above the nipple line, in the medial quadrant, in a slightly different location than was shown in her family doctor’s drawing. Dr. Roupenian recommended a biopsy, and on January 8, 1982, he performed the biopsy. When he could not aspirate any fluid from the mass with a needle, he removed the mass surgically. The pathologist reported the mass to be a benign fibroadenoma. Dr. Roupenian examined Policelli again on January 15, 1982. The incision was healing with slight drainage. Dr. Roupenian did not schedule or recommend any further follow-up visits.

Dr. Roupenian next saw Policelli on May 11, 1983, when she consulted him complaining of new lumps under her right arm. Dr. Roupenian found on examination that her entire right breast was hard and there were two nodes in the right armpit. He diagnosed a hematoma and recommended hot soaks. He scheduled a follow-up visit on June 10, 1983. At the June visit, he did not change his diagnosis or his recommended treatment.

Policelli did not return to Dr. Roupenian after her June visit. On June 21, 1983, she consulted Dr. Lawrence Geoghegan, a surgeon at Mount Auburn Hospital. Dr. Geoghegan immediately made a clinical diagnosis of advanced breast cancer and arranged for a biopsy and mammogram. He referred her to Dr. Roger Lange, an oncologist. Dr. Lange concurred in the diagnosis of advanced breast cancer. The tests confirmed the diagnosis. Dr. Lange treated Policelli with radiation and chemotherapy, and she did well for a short period. In March, 1984, however, tests revealed that the cancer had metastasized to her skeleton. Despite continuing treatment, Policelli died of cancer on May 8, 1986.

Policelli filed suit against Dr. Roupenian in October, 1984, claiming that Dr. Roupenian had been negligent in failing to diagnose cancer at the time of the biopsy, and in failing to *28 provide adequate follow-up and diagnostic care in the succeeding months.

Dr. Lange, Policelli’s oncologist, and Dr. Geoghegan, her surgeon, both testified on Policelli’s behalf. Two physicians who had not treated her, Dr. Guy Robbins, a surgical oncologist, and Dr. Norman Sadowsky, a radiologist, also testified as expert witnesses on her behalf. Dr. Lange described the stages of breast cancer and the way that “clinical doubling time,” or the growth rate of cancers can be determined or estimated. He also described the stages of cancer he observed in Policelli, and the growth rate of that cancer. The plaintiffs’ counsel then asked Dr. Lange, “Do you have an opinion with a reasonable degree of medical certainty as to whether the cancer which you have defined was present in January, 1982, was likewise diagnosable?” Defense counsel objected that there was insufficient basis for the question, and that it was inappropriately phrased because “anything is possible.” At a sidebar conference, the plaintiffs’ counsel offered to rephrase the question to ask whether “the average qualified surgeon in 1982 would have diagnosed this.” The judge responded that “[ejven then, it’s kind of far-fetched,” and sustained the defendant’s objection. The exclusion of that opinion evidence is the main focus of the plaintiffs’ appeal.

“The role of an expert witness is to help the jury understand issues of fact beyond their common experience. Under modern standards, expert testimony on matters within the witness’s field of expertise is admissible whenever it will aid the jury in reaching a decision, even if the expert’s opinion touches on the ultimate issues that the jury must decide. Commonwealth v. LaCorte, 373 Mass. 700, 705 (1977). Commonwealth v. Montmeny, 360 Mass. 526, 527-528 (1971).” Simony v. Solomon, 385 Mass. 91, 105 (1982). “Expert opinion, particularly when addressed to a jury, must be based on either the expert’s direct personal knowledge, on evidence already in the record or which the parties represent will be presented during the course of the trial, or on a com *29 bination of these sources.” LaClair v. Silberline Mfg. Co., 379 Mass. 21, 32 (1979). 3

The facts on which Dr. Lange was to have based his opinion were clearly before the jury. The record shows that before posing the question that was excluded, the plaintiffs’ counsel asked Dr. Lange a series of questions concerning the development of tumors in general and his observations of Policelli’s tumor in particular. Dr. Lange testified that in his opinion, based on his personal observations, facts in evidence, and specialized knowledge, Policelli suffered from cancer of the right breast in January, 1982, and that the cancer was in stage one or stage zero. Dr. Lange said that Policelli suffered from an aggressive cancer, with a growth rate near “the faster end of the doubling time of breast cancer.” Dr. Lange explained the concept of doubling time, and that three months is an “extremely rapid” doubling time for breast cancer. Lange then stated that, based on his clinical observations of Policelli in June, 1983, and using a three-month doubling time, he could project backwards in time to estimate the size of the tumor in January, 1982. This testimony forms an ample foundation for the question which followed, asking Dr. Lange whether the average qualified surgeon in 1982 would have diagnosed the cancer. There was, therefore, a sufficient factual basis for the expert’s opinion. 4

The judge erred when he excluded this question as “kind of far-fetched,” because an adequate factual foundation had been laid. So long as the facts on which the doctor’s opinion *30 is based are in evidence, “[t]he question whether the basis of the doctor’s opinion is sound goes to the weight of the evidence, not its admissibility.” Baker v. Commercial Union Ins. Co., 382 Mass. 347, 351 (1981). An adverse party is free to attack inconsistencies or omissions in the factual foundation, as well as flaws in the expert’s analytical process. Any inconsistencies, omissions, or flaws may be explored through cross-examination to affect the weight of such evidence, but such inconsistencies, omissions, or flaws do not affect the admissibility of such evidence.

The defendant maintains that the evidence properly was excluded as speculative. He contends that Dr. Lange’s answer would have been nothing more than conjecture because he did not express an unequivocal opinion as to exactly what stage he believed the cancer to have been in when Policelli first consulted Dr. Roupenian: Dr. Lange testified that the cancer was either in stage one or stage zero in January, 1982. There is no evidence in the record which supports the defendant’s claim on appeal that it is necessary to know the exact stage of the cancer in order to infer whether the cancer was detectable. Absent the witness’s expert testimony that the answer to the question would have been different depending on which stage the cancer was in, any uncertainty about the stage of the cancer is not relevant to the admissibility of the question.

We agree with the defendant that a litigant is not entitled to a new trial unless the error has “injuriously affected” her “substantial rights.” G. L. c. 231, §§ 119, 132 (1988 ed.). “[T]he substantial rights of a party are adversely affected when relevant evidence is erroneously excluded that, viewing the record in a commonsense way, could have made a material difference.” DeJesus v. Yogel, 404 Mass. 44, 48 (1989).

We do not agree with the defendant that the error in this case was nonprejudicial. Dr. Lange had a combination of qualifications which differed from the other witnesses. Dr. Lange was the treating oncologist, and so had extensive personal knowledge of the patient, the progression of her disease, as well as specialized knowledge of the growth and de *31 velopment of tumors. Dr. Lange’s credentials and his expertise differed from Dr. Geoghegan’s, the treating surgeon, and from the surgical oncologist’s and the radiologist’s who testified as expert witnesses, but who did not treat Policelli. 5

The defendant’s reliance on Murphy v. Conway, 360 Mass. 746 (1972), is misplaced. In that case, because the decedent’s husband, had testified to the same effect, the judge excluded testimony from a friend that the decedent had a sore throat. The expert witness’s testimony excluded in the instant case was not simply evidence as to the symptoms suffered by the plaintiff, but rather required the expert to apply his personal and technical knowledge to Policelli’s symptoms to aid the jury in determining the earlier course of Policelli’s cancer. The value of the excluded evidence depends on the knowledge and the credibility of the witness. An opinion from a nontreating expert is not the equivalent of an opinion from the treating expert with different expertise. Last, it is difficult to conclude the exclusion of this evidence was not prejudicial in light of the fact that the judge permitted each of the defendant’s two experts to give an opinion that the cancer was not diagnosable in January, 1982.

The excluded evidence bore directly on one of the essential issues in the plaintiffs’ case. The plaintiffs were required to adduce expert opinion on that issue. See Haggerty v. McCarthy, 344 Mass. 136 (1962). The opinion that was excluded was from an expert who was qualified to offer opinion evidence. The defendant’s experts were permitted to testify to their opinions whether the cancer was diagnosable in 1982. On this record, we cannot conclude that the exclusion of opinion evidence by the plaintiffs’ expert did not make a ma *32 terial difference to the jury. The plaintiffs are entitled to a new trial. 6

So ordered.

On the result we reach, we need not consider the plaintiffs’ other claims of error.

We have modified that rule “by permitting an expert to base an opinion on facts or data not in evidence if the facts or data are independently admissible and are a permissible basis for an expert to consider in formulating an opinion.” Department of Youth Servs. v. A Juvenile, 398 Mass. 516, 531 (1986). See Proposed Mass. R. Evid. 703. This modification is not involved in the instant case.

The defendant did not challenge Dr. Lange’s qualifications. The record shows that Dr. Lange is board certified in internal medicine, did a fellowship in hematology and oncology, teaches hematology and -oncology to fellows as an assistant clinical professor of Harvard Medical School, and maintains an active clinical practice. He has devoted much of his medical training, teaching, and practice to the study and treatment of tumors.

The defendant contends that one of the nontreating doctors, a surgical oncologist, who testified for the plaintiffs gave substantially the same evidence that was excluded by the judge. That physician, however, was not the treating oncologist, and did not have the same experience or expertise as Dr. Lange. Further, that physician was testifying to the negligent follow-up care, not to whether the cancer was diagnosable in January, 1982, and his testimony was not in the form of an expert opinion.

The defendant contends that, if a new trial is ordered, it should be limited to the theory that he was negligent in failing to diagnose cancer in January, 1982. He asserts that the evidence in question has no relevance to the plaintiffs’ claim that he also was negligent in follow-up treatment, and that the plaintiffs are therefore not entitled to a new trial on that issue. The defendant concedes in his brief, however, that these two theories of liability were not clearly separated at trial. Credible evidence that the cancer was detectable in January, 1982, might have affected the jury’s evaluation of the sufficiency of Dr. Roupenian’s follow-up care. We do not consider the two theories of negligence to be issues so separate and distinct as to limit the new trial to only one of the plaintiffs’ theories.